Initial commit

README.md

@@ -0,0 +1,60 @@
+---
+license: other
+base_model: meta-llama/Llama-2-7b-chat-hf
+tags:
+- llama-factory
+- full
+- generated_from_trainer
+model-index:
+- name: train_2024-07-16-16-48-49_llama2_2
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# train_2024-07-16-16-48-49_llama2_2
+
+This model is a fine-tuned version of [meta-llama/Llama-2-7b-chat-hf](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) on the truth_train_0716_2 dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-06
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 128
+- total_eval_batch_size: 64
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 10
+- num_epochs: 5.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.42.3
+- Pytorch 2.3.0a0+ebedce2
+- Datasets 2.20.0
+- Tokenizers 0.19.1

all_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.903225806451613,
+    "num_input_tokens_seen": 1299120,
+    "total_flos": 5.150239379108659e+16,
+    "train_loss": 0.3450760219912857,
+    "train_runtime": 2142.6975,
+    "train_samples_per_second": 11.57,
+    "train_steps_per_second": 0.089
+}

checkpoint-190/config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.42.3",
+  "use_cache": false,
+  "vocab_size": 32000
+}

checkpoint-190/generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.42.3"
+}

checkpoint-190/global_step190/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4d11d618f1219df097c6c9a06cf8fd4756d595fe3a00536877b1903a5218d3f3
+size 10107631180

checkpoint-190/global_step190/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c94be31d5660eec230ea527203a408a5ca653348e87f55c701a41e9a4acb8f67
+size 10107631372

checkpoint-190/global_step190/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d334ce67a3e3c77de78122a9e114a2c75be607fb2e9052c4967de098d58b20bd
+size 10107631436

checkpoint-190/global_step190/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c0f89dcb4840043f1db0b581fb9de8736afb45a5e05602835453588e3207911
+size 10107631372

checkpoint-190/global_step190/bf16_zero_pp_rank_4_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9d78e61c0f9d7f23bb7f77c0ff85f4a875249710029ad0b61df93507f72e81ad
+size 10107631436

checkpoint-190/global_step190/bf16_zero_pp_rank_5_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0a020b4655a1d83ccbf1ac7316e537814ae23a3821cf73ec1fec85cfeede94ab
+size 10107631500

checkpoint-190/global_step190/bf16_zero_pp_rank_6_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14dcce04d260abddd1df2470510e4981f99fc9bdf9903b044a08967a38005f9c
+size 10107631436

checkpoint-190/global_step190/bf16_zero_pp_rank_7_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa9bf5b58ddce0c99079b4709e4631b266da3615d50d7765a4dcfd70710c5a45
+size 10107630988

checkpoint-190/global_step190/mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30cf4c57c697259ba6764e03b3ba962e8c8f9e32e4ce43e625c2e8e960db8c2b
+size 13476919288

checkpoint-190/latest

@@ -0,0 +1 @@
+global_step190

checkpoint-190/model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af6cb31a20c84677f5085cfa5e077f8442dc10afc0d91be75e4e150668712487
+size 4938985352

checkpoint-190/model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03c38c9aa8e45f6b755a47ad5ba015f1b7c2daf7c1459bdb1b15cb876344adc0
+size 4947390880

checkpoint-190/model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7175530932f0eafafa2a482d31b7f8df84aa9dd5f3b5e23dd168cdeb053363a7
+size 3590488816

checkpoint-190/model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "total_size": 13476831232
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00003-of-00003.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.norm.weight": "model-00003-of-00003.safetensors"
+  }
+}

checkpoint-190/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a0ef6f96a48e59aa52c4b471312c2a62378c19acc7ebbae839612b03a7d775a
+size 15984

checkpoint-190/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab11d533c0fdad46ea8b8e295ba5fdb705e078eeb88cc28f37d82913508766e9
+size 15984

checkpoint-190/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:615c168147e3465ce5bfab6da2ff4afc68566ce00ec0f0c6c9fc988038a58d0a
+size 15984

checkpoint-190/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79f71e8f8674ecaef9f8cdcbf7ac457a8b8ff15b12694ba2a2fffcb4b43f0f08
+size 15984

checkpoint-190/rng_state_4.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88cf6d674dab5545c300a55135f08ca935730a3d35e2c419fb0b333f19482c19
+size 15984

checkpoint-190/rng_state_5.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2754f2cd8824702f027870d93748b3c0491b0ecd30f1e3d8e937116b2be6151f
+size 15984

checkpoint-190/rng_state_6.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1385124ac55604598f45ea6e2d141f29456647d3e7c10d12ca64ec93d312be8d
+size 15984

checkpoint-190/rng_state_7.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:416538efaec7391fa8fe782fb15146b83e5612d9e1961292c34c53e964806873
+size 15984

checkpoint-190/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e610dd5e9f27fcc75487bd147a6e92e36d0c3d068f260ff6e94230dced07d21d
+size 1064

checkpoint-190/special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

checkpoint-190/tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

checkpoint-190/tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if loop.index0 == 0 and system_message is defined %}{% set content = '<<SYS>>\n' + system_message + '\n<</SYS>>\n\n' + message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ '<s>' + '[INST] ' + content + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

checkpoint-190/trainer_state.json

@@ -0,0 +1,1553 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.903225806451613,
+  "eval_steps": 500,
+  "global_step": 190,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.025806451612903226,
+      "grad_norm": 276.5378112792969,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 8.2514,
+      "num_input_tokens_seen": 6848,
+      "step": 1
+    },
+    {
+      "epoch": 0.05161290322580645,
+      "grad_norm": 294.3509521484375,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 8.2793,
+      "num_input_tokens_seen": 13776,
+      "step": 2
+    },
+    {
+      "epoch": 0.07741935483870968,
+      "grad_norm": 283.4303283691406,
+      "learning_rate": 1.5e-06,
+      "loss": 8.17,
+      "num_input_tokens_seen": 20672,
+      "step": 3
+    },
+    {
+      "epoch": 0.1032258064516129,
+      "grad_norm": 285.55255126953125,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 7.6197,
+      "num_input_tokens_seen": 27744,
+      "step": 4
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 271.7382507324219,
+      "learning_rate": 2.5e-06,
+      "loss": 6.9491,
+      "num_input_tokens_seen": 34624,
+      "step": 5
+    },
+    {
+      "epoch": 0.15483870967741936,
+      "grad_norm": 125.66632080078125,
+      "learning_rate": 3e-06,
+      "loss": 5.2054,
+      "num_input_tokens_seen": 41424,
+      "step": 6
+    },
+    {
+      "epoch": 0.18064516129032257,
+      "grad_norm": 113.50272369384766,
+      "learning_rate": 3.5e-06,
+      "loss": 4.8642,
+      "num_input_tokens_seen": 48304,
+      "step": 7
+    },
+    {
+      "epoch": 0.2064516129032258,
+      "grad_norm": 113.24736785888672,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 3.2874,
+      "num_input_tokens_seen": 55152,
+      "step": 8
+    },
+    {
+      "epoch": 0.23225806451612904,
+      "grad_norm": 117.07583618164062,
+      "learning_rate": 4.5e-06,
+      "loss": 2.631,
+      "num_input_tokens_seen": 61680,
+      "step": 9
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 113.77539825439453,
+      "learning_rate": 5e-06,
+      "loss": 0.6982,
+      "num_input_tokens_seen": 68480,
+      "step": 10
+    },
+    {
+      "epoch": 0.2838709677419355,
+      "grad_norm": 36.774471282958984,
+      "learning_rate": 4.9996192378909785e-06,
+      "loss": 0.3276,
+      "num_input_tokens_seen": 75296,
+      "step": 11
+    },
+    {
+      "epoch": 0.3096774193548387,
+      "grad_norm": 33.21468734741211,
+      "learning_rate": 4.99847706754774e-06,
+      "loss": 0.293,
+      "num_input_tokens_seen": 82288,
+      "step": 12
+    },
+    {
+      "epoch": 0.33548387096774196,
+      "grad_norm": 5.65634822845459,
+      "learning_rate": 4.9965738368864345e-06,
+      "loss": 0.2129,
+      "num_input_tokens_seen": 89088,
+      "step": 13
+    },
+    {
+      "epoch": 0.36129032258064514,
+      "grad_norm": 55.87510299682617,
+      "learning_rate": 4.993910125649561e-06,
+      "loss": 0.4712,
+      "num_input_tokens_seen": 95904,
+      "step": 14
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 31.815500259399414,
+      "learning_rate": 4.990486745229364e-06,
+      "loss": 0.235,
+      "num_input_tokens_seen": 102896,
+      "step": 15
+    },
+    {
+      "epoch": 0.4129032258064516,
+      "grad_norm": 10.451783180236816,
+      "learning_rate": 4.986304738420684e-06,
+      "loss": 0.202,
+      "num_input_tokens_seen": 109792,
+      "step": 16
+    },
+    {
+      "epoch": 0.43870967741935485,
+      "grad_norm": 21.014144897460938,
+      "learning_rate": 4.981365379103306e-06,
+      "loss": 0.1981,
+      "num_input_tokens_seen": 116528,
+      "step": 17
+    },
+    {
+      "epoch": 0.4645161290322581,
+      "grad_norm": 4.897561550140381,
+      "learning_rate": 4.975670171853926e-06,
+      "loss": 0.1517,
+      "num_input_tokens_seen": 123200,
+      "step": 18
+    },
+    {
+      "epoch": 0.49032258064516127,
+      "grad_norm": 37.270225524902344,
+      "learning_rate": 4.9692208514878445e-06,
+      "loss": 0.4335,
+      "num_input_tokens_seen": 130032,
+      "step": 19
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 31.595401763916016,
+      "learning_rate": 4.962019382530521e-06,
+      "loss": 0.3609,
+      "num_input_tokens_seen": 136832,
+      "step": 20
+    },
+    {
+      "epoch": 0.5419354838709678,
+      "grad_norm": 8.859624862670898,
+      "learning_rate": 4.9540679586191605e-06,
+      "loss": 0.1708,
+      "num_input_tokens_seen": 143904,
+      "step": 21
+    },
+    {
+      "epoch": 0.567741935483871,
+      "grad_norm": 21.666885375976562,
+      "learning_rate": 4.9453690018345144e-06,
+      "loss": 0.2277,
+      "num_input_tokens_seen": 150496,
+      "step": 22
+    },
+    {
+      "epoch": 0.5935483870967742,
+      "grad_norm": 29.53874969482422,
+      "learning_rate": 4.935925161963089e-06,
+      "loss": 0.3437,
+      "num_input_tokens_seen": 157360,
+      "step": 23
+    },
+    {
+      "epoch": 0.6193548387096774,
+      "grad_norm": 17.89641761779785,
+      "learning_rate": 4.925739315689991e-06,
+      "loss": 0.2229,
+      "num_input_tokens_seen": 164464,
+      "step": 24
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 2.7967522144317627,
+      "learning_rate": 4.914814565722671e-06,
+      "loss": 0.1242,
+      "num_input_tokens_seen": 170960,
+      "step": 25
+    },
+    {
+      "epoch": 0.6709677419354839,
+      "grad_norm": 15.343024253845215,
+      "learning_rate": 4.903154239845798e-06,
+      "loss": 0.2117,
+      "num_input_tokens_seen": 177808,
+      "step": 26
+    },
+    {
+      "epoch": 0.6967741935483871,
+      "grad_norm": 20.621328353881836,
+      "learning_rate": 4.890761889907589e-06,
+      "loss": 0.2706,
+      "num_input_tokens_seen": 184848,
+      "step": 27
+    },
+    {
+      "epoch": 0.7225806451612903,
+      "grad_norm": 14.205361366271973,
+      "learning_rate": 4.8776412907378845e-06,
+      "loss": 0.2084,
+      "num_input_tokens_seen": 191536,
+      "step": 28
+    },
+    {
+      "epoch": 0.7483870967741936,
+      "grad_norm": 4.756993770599365,
+      "learning_rate": 4.863796438998293e-06,
+      "loss": 0.0981,
+      "num_input_tokens_seen": 198368,
+      "step": 29
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 13.791275024414062,
+      "learning_rate": 4.849231551964771e-06,
+      "loss": 0.16,
+      "num_input_tokens_seen": 205200,
+      "step": 30
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 11.285390853881836,
+      "learning_rate": 4.833951066243004e-06,
+      "loss": 0.1614,
+      "num_input_tokens_seen": 211760,
+      "step": 31
+    },
+    {
+      "epoch": 0.8258064516129032,
+      "grad_norm": 13.351017951965332,
+      "learning_rate": 4.817959636416969e-06,
+      "loss": 0.1742,
+      "num_input_tokens_seen": 218448,
+      "step": 32
+    },
+    {
+      "epoch": 0.8516129032258064,
+      "grad_norm": 6.028775691986084,
+      "learning_rate": 4.801262133631101e-06,
+      "loss": 0.1107,
+      "num_input_tokens_seen": 225408,
+      "step": 33
+    },
+    {
+      "epoch": 0.8774193548387097,
+      "grad_norm": 2.5520031452178955,
+      "learning_rate": 4.783863644106502e-06,
+      "loss": 0.0822,
+      "num_input_tokens_seen": 232032,
+      "step": 34
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 13.276092529296875,
+      "learning_rate": 4.765769467591626e-06,
+      "loss": 0.1873,
+      "num_input_tokens_seen": 238672,
+      "step": 35
+    },
+    {
+      "epoch": 0.9290322580645162,
+      "grad_norm": 15.978204727172852,
+      "learning_rate": 4.746985115747918e-06,
+      "loss": 0.2375,
+      "num_input_tokens_seen": 245456,
+      "step": 36
+    },
+    {
+      "epoch": 0.9548387096774194,
+      "grad_norm": 16.432758331298828,
+      "learning_rate": 4.72751631047092e-06,
+      "loss": 0.2667,
+      "num_input_tokens_seen": 252576,
+      "step": 37
+    },
+    {
+      "epoch": 0.9806451612903225,
+      "grad_norm": 12.175621032714844,
+      "learning_rate": 4.707368982147318e-06,
+      "loss": 0.1547,
+      "num_input_tokens_seen": 259536,
+      "step": 38
+    },
+    {
+      "epoch": 1.0064516129032257,
+      "grad_norm": 9.780308723449707,
+      "learning_rate": 4.68654926784849e-06,
+      "loss": 0.1662,
+      "num_input_tokens_seen": 266560,
+      "step": 39
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 3.619741678237915,
+      "learning_rate": 4.665063509461098e-06,
+      "loss": 0.0808,
+      "num_input_tokens_seen": 273248,
+      "step": 40
+    },
+    {
+      "epoch": 1.0580645161290323,
+      "grad_norm": 5.894082546234131,
+      "learning_rate": 4.642918251755281e-06,
+      "loss": 0.0884,
+      "num_input_tokens_seen": 279888,
+      "step": 41
+    },
+    {
+      "epoch": 1.0838709677419356,
+      "grad_norm": 5.443681240081787,
+      "learning_rate": 4.620120240391065e-06,
+      "loss": 0.0883,
+      "num_input_tokens_seen": 286944,
+      "step": 42
+    },
+    {
+      "epoch": 1.1096774193548387,
+      "grad_norm": 3.3760452270507812,
+      "learning_rate": 4.596676419863561e-06,
+      "loss": 0.0562,
+      "num_input_tokens_seen": 293792,
+      "step": 43
+    },
+    {
+      "epoch": 1.135483870967742,
+      "grad_norm": 2.7039427757263184,
+      "learning_rate": 4.572593931387604e-06,
+      "loss": 0.0856,
+      "num_input_tokens_seen": 300512,
+      "step": 44
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 4.186523914337158,
+      "learning_rate": 4.54788011072248e-06,
+      "loss": 0.0612,
+      "num_input_tokens_seen": 307264,
+      "step": 45
+    },
+    {
+      "epoch": 1.1870967741935483,
+      "grad_norm": 4.6027374267578125,
+      "learning_rate": 4.522542485937369e-06,
+      "loss": 0.0944,
+      "num_input_tokens_seen": 314112,
+      "step": 46
+    },
+    {
+      "epoch": 1.2129032258064516,
+      "grad_norm": 4.636350154876709,
+      "learning_rate": 4.496588775118232e-06,
+      "loss": 0.0624,
+      "num_input_tokens_seen": 320864,
+      "step": 47
+    },
+    {
+      "epoch": 1.238709677419355,
+      "grad_norm": 3.0458312034606934,
+      "learning_rate": 4.470026884016805e-06,
+      "loss": 0.0363,
+      "num_input_tokens_seen": 327648,
+      "step": 48
+    },
+    {
+      "epoch": 1.2645161290322582,
+      "grad_norm": 4.572902679443359,
+      "learning_rate": 4.442864903642428e-06,
+      "loss": 0.1039,
+      "num_input_tokens_seen": 334432,
+      "step": 49
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 2.629575252532959,
+      "learning_rate": 4.415111107797445e-06,
+      "loss": 0.0488,
+      "num_input_tokens_seen": 341248,
+      "step": 50
+    },
+    {
+      "epoch": 1.3161290322580645,
+      "grad_norm": 2.7692055702209473,
+      "learning_rate": 4.386773950556931e-06,
+      "loss": 0.0613,
+      "num_input_tokens_seen": 348192,
+      "step": 51
+    },
+    {
+      "epoch": 1.3419354838709676,
+      "grad_norm": 2.706336498260498,
+      "learning_rate": 4.357862063693486e-06,
+      "loss": 0.07,
+      "num_input_tokens_seen": 354816,
+      "step": 52
+    },
+    {
+      "epoch": 1.367741935483871,
+      "grad_norm": 1.679677963256836,
+      "learning_rate": 4.328384254047927e-06,
+      "loss": 0.0463,
+      "num_input_tokens_seen": 361600,
+      "step": 53
+    },
+    {
+      "epoch": 1.3935483870967742,
+      "grad_norm": 2.623893976211548,
+      "learning_rate": 4.2983495008466285e-06,
+      "loss": 0.0671,
+      "num_input_tokens_seen": 368464,
+      "step": 54
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 1.7314574718475342,
+      "learning_rate": 4.267766952966369e-06,
+      "loss": 0.0428,
+      "num_input_tokens_seen": 375440,
+      "step": 55
+    },
+    {
+      "epoch": 1.4451612903225808,
+      "grad_norm": 5.240540981292725,
+      "learning_rate": 4.236645926147493e-06,
+      "loss": 0.0678,
+      "num_input_tokens_seen": 382240,
+      "step": 56
+    },
+    {
+      "epoch": 1.4709677419354839,
+      "grad_norm": 1.9007383584976196,
+      "learning_rate": 4.204995900156247e-06,
+      "loss": 0.0476,
+      "num_input_tokens_seen": 389008,
+      "step": 57
+    },
+    {
+      "epoch": 1.4967741935483871,
+      "grad_norm": 3.100925922393799,
+      "learning_rate": 4.172826515897146e-06,
+      "loss": 0.0442,
+      "num_input_tokens_seen": 396080,
+      "step": 58
+    },
+    {
+      "epoch": 1.5225806451612902,
+      "grad_norm": 2.798711061477661,
+      "learning_rate": 4.140147572476269e-06,
+      "loss": 0.0336,
+      "num_input_tokens_seen": 403056,
+      "step": 59
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 2.20042085647583,
+      "learning_rate": 4.106969024216348e-06,
+      "loss": 0.046,
+      "num_input_tokens_seen": 409664,
+      "step": 60
+    },
+    {
+      "epoch": 1.5741935483870968,
+      "grad_norm": 4.055737495422363,
+      "learning_rate": 4.073300977624594e-06,
+      "loss": 0.0416,
+      "num_input_tokens_seen": 416272,
+      "step": 61
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 3.15283465385437,
+      "learning_rate": 4.039153688314146e-06,
+      "loss": 0.0649,
+      "num_input_tokens_seen": 423360,
+      "step": 62
+    },
+    {
+      "epoch": 1.6258064516129034,
+      "grad_norm": 4.418552398681641,
+      "learning_rate": 4.0045375578801216e-06,
+      "loss": 0.0591,
+      "num_input_tokens_seen": 430304,
+      "step": 63
+    },
+    {
+      "epoch": 1.6516129032258065,
+      "grad_norm": 3.960134506225586,
+      "learning_rate": 3.969463130731183e-06,
+      "loss": 0.0318,
+      "num_input_tokens_seen": 436960,
+      "step": 64
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 3.1769161224365234,
+      "learning_rate": 3.933941090877615e-06,
+      "loss": 0.0462,
+      "num_input_tokens_seen": 443792,
+      "step": 65
+    },
+    {
+      "epoch": 1.7032258064516128,
+      "grad_norm": 4.235939025878906,
+      "learning_rate": 3.897982258676867e-06,
+      "loss": 0.0465,
+      "num_input_tokens_seen": 450672,
+      "step": 66
+    },
+    {
+      "epoch": 1.729032258064516,
+      "grad_norm": 2.7547543048858643,
+      "learning_rate": 3.861597587537568e-06,
+      "loss": 0.0316,
+      "num_input_tokens_seen": 457616,
+      "step": 67
+    },
+    {
+      "epoch": 1.7548387096774194,
+      "grad_norm": 4.103342533111572,
+      "learning_rate": 3.824798160583012e-06,
+      "loss": 0.1,
+      "num_input_tokens_seen": 464592,
+      "step": 68
+    },
+    {
+      "epoch": 1.7806451612903227,
+      "grad_norm": 3.525808811187744,
+      "learning_rate": 3.787595187275136e-06,
+      "loss": 0.0711,
+      "num_input_tokens_seen": 471328,
+      "step": 69
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 2.9660446643829346,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.0494,
+      "num_input_tokens_seen": 478480,
+      "step": 70
+    },
+    {
+      "epoch": 1.832258064516129,
+      "grad_norm": 3.661849021911621,
+      "learning_rate": 3.7120240506158433e-06,
+      "loss": 0.0618,
+      "num_input_tokens_seen": 485264,
+      "step": 71
+    },
+    {
+      "epoch": 1.8580645161290321,
+      "grad_norm": 1.8227900266647339,
+      "learning_rate": 3.6736789069647273e-06,
+      "loss": 0.0511,
+      "num_input_tokens_seen": 492064,
+      "step": 72
+    },
+    {
+      "epoch": 1.8838709677419354,
+      "grad_norm": 2.1322553157806396,
+      "learning_rate": 3.634976249348867e-06,
+      "loss": 0.0464,
+      "num_input_tokens_seen": 498896,
+      "step": 73
+    },
+    {
+      "epoch": 1.9096774193548387,
+      "grad_norm": 1.8635531663894653,
+      "learning_rate": 3.595927866972694e-06,
+      "loss": 0.0331,
+      "num_input_tokens_seen": 506016,
+      "step": 74
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 2.9526419639587402,
+      "learning_rate": 3.556545654351749e-06,
+      "loss": 0.0706,
+      "num_input_tokens_seen": 512848,
+      "step": 75
+    },
+    {
+      "epoch": 1.9612903225806453,
+      "grad_norm": 2.8351802825927734,
+      "learning_rate": 3.516841607689501e-06,
+      "loss": 0.0442,
+      "num_input_tokens_seen": 519744,
+      "step": 76
+    },
+    {
+      "epoch": 1.9870967741935484,
+      "grad_norm": 3.1449944972991943,
+      "learning_rate": 3.476827821223184e-06,
+      "loss": 0.042,
+      "num_input_tokens_seen": 526352,
+      "step": 77
+    },
+    {
+      "epoch": 2.0129032258064514,
+      "grad_norm": 1.6314219236373901,
+      "learning_rate": 3.436516483539781e-06,
+      "loss": 0.021,
+      "num_input_tokens_seen": 533136,
+      "step": 78
+    },
+    {
+      "epoch": 2.0387096774193547,
+      "grad_norm": 1.6224279403686523,
+      "learning_rate": 3.39591987386325e-06,
+      "loss": 0.0094,
+      "num_input_tokens_seen": 540016,
+      "step": 79
+    },
+    {
+      "epoch": 2.064516129032258,
+      "grad_norm": 0.3502344787120819,
+      "learning_rate": 3.3550503583141726e-06,
+      "loss": 0.0021,
+      "num_input_tokens_seen": 547152,
+      "step": 80
+    },
+    {
+      "epoch": 2.0903225806451613,
+      "grad_norm": 1.7473047971725464,
+      "learning_rate": 3.313920386142892e-06,
+      "loss": 0.0146,
+      "num_input_tokens_seen": 553888,
+      "step": 81
+    },
+    {
+      "epoch": 2.1161290322580646,
+      "grad_norm": 2.363898277282715,
+      "learning_rate": 3.272542485937369e-06,
+      "loss": 0.0237,
+      "num_input_tokens_seen": 561024,
+      "step": 82
+    },
+    {
+      "epoch": 2.141935483870968,
+      "grad_norm": 1.377082109451294,
+      "learning_rate": 3.230929261806842e-06,
+      "loss": 0.0031,
+      "num_input_tokens_seen": 567936,
+      "step": 83
+    },
+    {
+      "epoch": 2.167741935483871,
+      "grad_norm": 0.6467981934547424,
+      "learning_rate": 3.189093389542498e-06,
+      "loss": 0.0034,
+      "num_input_tokens_seen": 574736,
+      "step": 84
+    },
+    {
+      "epoch": 2.193548387096774,
+      "grad_norm": 0.9653486609458923,
+      "learning_rate": 3.147047612756302e-06,
+      "loss": 0.0045,
+      "num_input_tokens_seen": 581680,
+      "step": 85
+    },
+    {
+      "epoch": 2.2193548387096773,
+      "grad_norm": 0.7019590735435486,
+      "learning_rate": 3.1048047389991693e-06,
+      "loss": 0.0031,
+      "num_input_tokens_seen": 588464,
+      "step": 86
+    },
+    {
+      "epoch": 2.2451612903225806,
+      "grad_norm": 2.078730821609497,
+      "learning_rate": 3.062377635859663e-06,
+      "loss": 0.0341,
+      "num_input_tokens_seen": 595312,
+      "step": 87
+    },
+    {
+      "epoch": 2.270967741935484,
+      "grad_norm": 4.608306407928467,
+      "learning_rate": 3.019779227044398e-06,
+      "loss": 0.0095,
+      "num_input_tokens_seen": 602080,
+      "step": 88
+    },
+    {
+      "epoch": 2.296774193548387,
+      "grad_norm": 3.8219709396362305,
+      "learning_rate": 2.9770224884413625e-06,
+      "loss": 0.0459,
+      "num_input_tokens_seen": 609168,
+      "step": 89
+    },
+    {
+      "epoch": 2.3225806451612905,
+      "grad_norm": 2.868053674697876,
+      "learning_rate": 2.9341204441673267e-06,
+      "loss": 0.0104,
+      "num_input_tokens_seen": 616000,
+      "step": 90
+    },
+    {
+      "epoch": 2.3483870967741938,
+      "grad_norm": 2.516240119934082,
+      "learning_rate": 2.8910861626005774e-06,
+      "loss": 0.0201,
+      "num_input_tokens_seen": 622544,
+      "step": 91
+    },
+    {
+      "epoch": 2.3741935483870966,
+      "grad_norm": 0.5750585794448853,
+      "learning_rate": 2.847932752400164e-06,
+      "loss": 0.0021,
+      "num_input_tokens_seen": 629600,
+      "step": 92
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 2.9817891120910645,
+      "learning_rate": 2.804673358512869e-06,
+      "loss": 0.043,
+      "num_input_tokens_seen": 636320,
+      "step": 93
+    },
+    {
+      "epoch": 2.425806451612903,
+      "grad_norm": 1.8277342319488525,
+      "learning_rate": 2.761321158169134e-06,
+      "loss": 0.0207,
+      "num_input_tokens_seen": 643136,
+      "step": 94
+    },
+    {
+      "epoch": 2.4516129032258065,
+      "grad_norm": 4.615224361419678,
+      "learning_rate": 2.717889356869146e-06,
+      "loss": 0.0148,
+      "num_input_tokens_seen": 650048,
+      "step": 95
+    },
+    {
+      "epoch": 2.47741935483871,
+      "grad_norm": 0.9854787588119507,
+      "learning_rate": 2.6743911843603134e-06,
+      "loss": 0.004,
+      "num_input_tokens_seen": 656736,
+      "step": 96
+    },
+    {
+      "epoch": 2.5032258064516126,
+      "grad_norm": 1.4292289018630981,
+      "learning_rate": 2.6308398906073603e-06,
+      "loss": 0.0131,
+      "num_input_tokens_seen": 663552,
+      "step": 97
+    },
+    {
+      "epoch": 2.5290322580645164,
+      "grad_norm": 4.594967842102051,
+      "learning_rate": 2.587248741756253e-06,
+      "loss": 0.0455,
+      "num_input_tokens_seen": 670224,
+      "step": 98
+    },
+    {
+      "epoch": 2.554838709677419,
+      "grad_norm": 0.8579672574996948,
+      "learning_rate": 2.543631016093209e-06,
+      "loss": 0.0031,
+      "num_input_tokens_seen": 676960,
+      "step": 99
+    },
+    {
+      "epoch": 2.5806451612903225,
+      "grad_norm": 2.075626850128174,
+      "learning_rate": 2.5e-06,
+      "loss": 0.0099,
+      "num_input_tokens_seen": 684016,
+      "step": 100
+    },
+    {
+      "epoch": 2.606451612903226,
+      "grad_norm": 3.9176177978515625,
+      "learning_rate": 2.4563689839067913e-06,
+      "loss": 0.0797,
+      "num_input_tokens_seen": 690832,
+      "step": 101
+    },
+    {
+      "epoch": 2.632258064516129,
+      "grad_norm": 1.6330746412277222,
+      "learning_rate": 2.4127512582437486e-06,
+      "loss": 0.0059,
+      "num_input_tokens_seen": 697808,
+      "step": 102
+    },
+    {
+      "epoch": 2.6580645161290324,
+      "grad_norm": 2.173410177230835,
+      "learning_rate": 2.3691601093926406e-06,
+      "loss": 0.0438,
+      "num_input_tokens_seen": 704352,
+      "step": 103
+    },
+    {
+      "epoch": 2.6838709677419352,
+      "grad_norm": 1.6522724628448486,
+      "learning_rate": 2.325608815639687e-06,
+      "loss": 0.0149,
+      "num_input_tokens_seen": 711456,
+      "step": 104
+    },
+    {
+      "epoch": 2.709677419354839,
+      "grad_norm": 0.7776157855987549,
+      "learning_rate": 2.2821106431308546e-06,
+      "loss": 0.0126,
+      "num_input_tokens_seen": 718224,
+      "step": 105
+    },
+    {
+      "epoch": 2.735483870967742,
+      "grad_norm": 2.2912323474884033,
+      "learning_rate": 2.238678841830867e-06,
+      "loss": 0.0255,
+      "num_input_tokens_seen": 724928,
+      "step": 106
+    },
+    {
+      "epoch": 2.761290322580645,
+      "grad_norm": 0.5118953585624695,
+      "learning_rate": 2.195326641487132e-06,
+      "loss": 0.0048,
+      "num_input_tokens_seen": 731616,
+      "step": 107
+    },
+    {
+      "epoch": 2.7870967741935484,
+      "grad_norm": 0.9067234396934509,
+      "learning_rate": 2.1520672475998374e-06,
+      "loss": 0.0142,
+      "num_input_tokens_seen": 738272,
+      "step": 108
+    },
+    {
+      "epoch": 2.8129032258064517,
+      "grad_norm": 1.6885769367218018,
+      "learning_rate": 2.1089138373994226e-06,
+      "loss": 0.0193,
+      "num_input_tokens_seen": 745104,
+      "step": 109
+    },
+    {
+      "epoch": 2.838709677419355,
+      "grad_norm": 0.6433587074279785,
+      "learning_rate": 2.0658795558326745e-06,
+      "loss": 0.0055,
+      "num_input_tokens_seen": 752016,
+      "step": 110
+    },
+    {
+      "epoch": 2.864516129032258,
+      "grad_norm": 1.0263943672180176,
+      "learning_rate": 2.022977511558638e-06,
+      "loss": 0.0144,
+      "num_input_tokens_seen": 758720,
+      "step": 111
+    },
+    {
+      "epoch": 2.8903225806451616,
+      "grad_norm": 1.3137305974960327,
+      "learning_rate": 1.9802207729556023e-06,
+      "loss": 0.0272,
+      "num_input_tokens_seen": 765520,
+      "step": 112
+    },
+    {
+      "epoch": 2.9161290322580644,
+      "grad_norm": 2.692868947982788,
+      "learning_rate": 1.937622364140338e-06,
+      "loss": 0.0101,
+      "num_input_tokens_seen": 772288,
+      "step": 113
+    },
+    {
+      "epoch": 2.9419354838709677,
+      "grad_norm": 2.19588565826416,
+      "learning_rate": 1.895195261000831e-06,
+      "loss": 0.0109,
+      "num_input_tokens_seen": 779520,
+      "step": 114
+    },
+    {
+      "epoch": 2.967741935483871,
+      "grad_norm": 2.681143283843994,
+      "learning_rate": 1.852952387243698e-06,
+      "loss": 0.018,
+      "num_input_tokens_seen": 786400,
+      "step": 115
+    },
+    {
+      "epoch": 2.9935483870967743,
+      "grad_norm": 1.9662718772888184,
+      "learning_rate": 1.8109066104575023e-06,
+      "loss": 0.0141,
+      "num_input_tokens_seen": 793008,
+      "step": 116
+    },
+    {
+      "epoch": 3.0193548387096776,
+      "grad_norm": 1.1327028274536133,
+      "learning_rate": 1.7690707381931585e-06,
+      "loss": 0.0057,
+      "num_input_tokens_seen": 799808,
+      "step": 117
+    },
+    {
+      "epoch": 3.0451612903225804,
+      "grad_norm": 0.593132495880127,
+      "learning_rate": 1.7274575140626318e-06,
+      "loss": 0.0063,
+      "num_input_tokens_seen": 806576,
+      "step": 118
+    },
+    {
+      "epoch": 3.0709677419354837,
+      "grad_norm": 1.123922348022461,
+      "learning_rate": 1.686079613857109e-06,
+      "loss": 0.0138,
+      "num_input_tokens_seen": 813312,
+      "step": 119
+    },
+    {
+      "epoch": 3.096774193548387,
+      "grad_norm": 0.14445500075817108,
+      "learning_rate": 1.6449496416858285e-06,
+      "loss": 0.0011,
+      "num_input_tokens_seen": 819728,
+      "step": 120
+    },
+    {
+      "epoch": 3.1225806451612903,
+      "grad_norm": 0.09276027232408524,
+      "learning_rate": 1.6040801261367494e-06,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 826496,
+      "step": 121
+    },
+    {
+      "epoch": 3.1483870967741936,
+      "grad_norm": 0.6197965741157532,
+      "learning_rate": 1.56348351646022e-06,
+      "loss": 0.0055,
+      "num_input_tokens_seen": 833360,
+      "step": 122
+    },
+    {
+      "epoch": 3.174193548387097,
+      "grad_norm": 0.1872132420539856,
+      "learning_rate": 1.5231721787768162e-06,
+      "loss": 0.0011,
+      "num_input_tokens_seen": 840080,
+      "step": 123
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 2.89666748046875,
+      "learning_rate": 1.4831583923105e-06,
+      "loss": 0.0173,
+      "num_input_tokens_seen": 847104,
+      "step": 124
+    },
+    {
+      "epoch": 3.225806451612903,
+      "grad_norm": 0.8067394495010376,
+      "learning_rate": 1.443454345648252e-06,
+      "loss": 0.0027,
+      "num_input_tokens_seen": 853712,
+      "step": 125
+    },
+    {
+      "epoch": 3.2516129032258063,
+      "grad_norm": 0.5419718623161316,
+      "learning_rate": 1.4040721330273063e-06,
+      "loss": 0.0029,
+      "num_input_tokens_seen": 860400,
+      "step": 126
+    },
+    {
+      "epoch": 3.2774193548387096,
+      "grad_norm": 0.05267687141895294,
+      "learning_rate": 1.3650237506511333e-06,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 867440,
+      "step": 127
+    },
+    {
+      "epoch": 3.303225806451613,
+      "grad_norm": 0.11772596091032028,
+      "learning_rate": 1.3263210930352737e-06,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 874256,
+      "step": 128
+    },
+    {
+      "epoch": 3.329032258064516,
+      "grad_norm": 0.9182859063148499,
+      "learning_rate": 1.2879759493841577e-06,
+      "loss": 0.008,
+      "num_input_tokens_seen": 881168,
+      "step": 129
+    },
+    {
+      "epoch": 3.3548387096774195,
+      "grad_norm": 0.07012775540351868,
+      "learning_rate": 1.2500000000000007e-06,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 888128,
+      "step": 130
+    },
+    {
+      "epoch": 3.3806451612903228,
+      "grad_norm": 0.979766309261322,
+      "learning_rate": 1.2124048127248644e-06,
+      "loss": 0.0049,
+      "num_input_tokens_seen": 895296,
+      "step": 131
+    },
+    {
+      "epoch": 3.4064516129032256,
+      "grad_norm": 0.25211745500564575,
+      "learning_rate": 1.1752018394169882e-06,
+      "loss": 0.0012,
+      "num_input_tokens_seen": 901984,
+      "step": 132
+    },
+    {
+      "epoch": 3.432258064516129,
+      "grad_norm": 0.7576802968978882,
+      "learning_rate": 1.1384024124624324e-06,
+      "loss": 0.0044,
+      "num_input_tokens_seen": 909104,
+      "step": 133
+    },
+    {
+      "epoch": 3.458064516129032,
+      "grad_norm": 0.6242833733558655,
+      "learning_rate": 1.1020177413231334e-06,
+      "loss": 0.0017,
+      "num_input_tokens_seen": 916032,
+      "step": 134
+    },
+    {
+      "epoch": 3.4838709677419355,
+      "grad_norm": 0.06409750878810883,
+      "learning_rate": 1.0660589091223854e-06,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 922736,
+      "step": 135
+    },
+    {
+      "epoch": 3.509677419354839,
+      "grad_norm": 0.7863635420799255,
+      "learning_rate": 1.0305368692688175e-06,
+      "loss": 0.0099,
+      "num_input_tokens_seen": 929472,
+      "step": 136
+    },
+    {
+      "epoch": 3.535483870967742,
+      "grad_norm": 0.7082129716873169,
+      "learning_rate": 9.95462442119879e-07,
+      "loss": 0.0068,
+      "num_input_tokens_seen": 936112,
+      "step": 137
+    },
+    {
+      "epoch": 3.5612903225806454,
+      "grad_norm": 0.5010254383087158,
+      "learning_rate": 9.608463116858544e-07,
+      "loss": 0.0025,
+      "num_input_tokens_seen": 942912,
+      "step": 138
+    },
+    {
+      "epoch": 3.587096774193548,
+      "grad_norm": 0.1011819988489151,
+      "learning_rate": 9.266990223754069e-07,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 949984,
+      "step": 139
+    },
+    {
+      "epoch": 3.6129032258064515,
+      "grad_norm": 4.317319869995117,
+      "learning_rate": 8.930309757836517e-07,
+      "loss": 0.0101,
+      "num_input_tokens_seen": 956608,
+      "step": 140
+    },
+    {
+      "epoch": 3.638709677419355,
+      "grad_norm": 0.7055479288101196,
+      "learning_rate": 8.598524275237321e-07,
+      "loss": 0.0068,
+      "num_input_tokens_seen": 963488,
+      "step": 141
+    },
+    {
+      "epoch": 3.664516129032258,
+      "grad_norm": 0.10063151270151138,
+      "learning_rate": 8.271734841028553e-07,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 970128,
+      "step": 142
+    },
+    {
+      "epoch": 3.6903225806451614,
+      "grad_norm": 3.3077032566070557,
+      "learning_rate": 7.950040998437541e-07,
+      "loss": 0.0161,
+      "num_input_tokens_seen": 976800,
+      "step": 143
+    },
+    {
+      "epoch": 3.7161290322580647,
+      "grad_norm": 1.0670413970947266,
+      "learning_rate": 7.633540738525066e-07,
+      "loss": 0.0115,
+      "num_input_tokens_seen": 984000,
+      "step": 144
+    },
+    {
+      "epoch": 3.741935483870968,
+      "grad_norm": 1.591837763786316,
+      "learning_rate": 7.322330470336314e-07,
+      "loss": 0.0052,
+      "num_input_tokens_seen": 990896,
+      "step": 145
+    },
+    {
+      "epoch": 3.767741935483871,
+      "grad_norm": 3.105447769165039,
+      "learning_rate": 7.016504991533727e-07,
+      "loss": 0.0098,
+      "num_input_tokens_seen": 997824,
+      "step": 146
+    },
+    {
+      "epoch": 3.793548387096774,
+      "grad_norm": 0.0832144096493721,
+      "learning_rate": 6.716157459520739e-07,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1004928,
+      "step": 147
+    },
+    {
+      "epoch": 3.8193548387096774,
+      "grad_norm": 0.2546854317188263,
+      "learning_rate": 6.421379363065142e-07,
+      "loss": 0.0012,
+      "num_input_tokens_seen": 1011696,
+      "step": 148
+    },
+    {
+      "epoch": 3.8451612903225807,
+      "grad_norm": 0.40274935960769653,
+      "learning_rate": 6.1322604944307e-07,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 1018368,
+      "step": 149
+    },
+    {
+      "epoch": 3.870967741935484,
+      "grad_norm": 0.04935682937502861,
+      "learning_rate": 5.848888922025553e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1025424,
+      "step": 150
+    },
+    {
+      "epoch": 3.896774193548387,
+      "grad_norm": 0.8725687265396118,
+      "learning_rate": 5.571350963575728e-07,
+      "loss": 0.0026,
+      "num_input_tokens_seen": 1032016,
+      "step": 151
+    },
+    {
+      "epoch": 3.9225806451612906,
+      "grad_norm": 0.9522215127944946,
+      "learning_rate": 5.299731159831953e-07,
+      "loss": 0.0097,
+      "num_input_tokens_seen": 1038928,
+      "step": 152
+    },
+    {
+      "epoch": 3.9483870967741934,
+      "grad_norm": 2.608211040496826,
+      "learning_rate": 5.034112248817685e-07,
+      "loss": 0.0047,
+      "num_input_tokens_seen": 1045664,
+      "step": 153
+    },
+    {
+      "epoch": 3.9741935483870967,
+      "grad_norm": 0.913472592830658,
+      "learning_rate": 4.774575140626317e-07,
+      "loss": 0.0081,
+      "num_input_tokens_seen": 1052640,
+      "step": 154
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.5549633502960205,
+      "learning_rate": 4.5211988927752026e-07,
+      "loss": 0.0018,
+      "num_input_tokens_seen": 1059840,
+      "step": 155
+    },
+    {
+      "epoch": 4.025806451612903,
+      "grad_norm": 0.6482688784599304,
+      "learning_rate": 4.27406068612396e-07,
+      "loss": 0.0053,
+      "num_input_tokens_seen": 1066704,
+      "step": 156
+    },
+    {
+      "epoch": 4.051612903225807,
+      "grad_norm": 0.11119811981916428,
+      "learning_rate": 4.033235801364402e-07,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1073600,
+      "step": 157
+    },
+    {
+      "epoch": 4.077419354838709,
+      "grad_norm": 0.009617508389055729,
+      "learning_rate": 3.798797596089351e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1080208,
+      "step": 158
+    },
+    {
+      "epoch": 4.103225806451613,
+      "grad_norm": 0.18060296773910522,
+      "learning_rate": 3.5708174824471947e-07,
+      "loss": 0.0018,
+      "num_input_tokens_seen": 1086880,
+      "step": 159
+    },
+    {
+      "epoch": 4.129032258064516,
+      "grad_norm": 0.3545215427875519,
+      "learning_rate": 3.3493649053890325e-07,
+      "loss": 0.001,
+      "num_input_tokens_seen": 1093520,
+      "step": 160
+    },
+    {
+      "epoch": 4.15483870967742,
+      "grad_norm": 0.03870348632335663,
+      "learning_rate": 3.134507321515107e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1100528,
+      "step": 161
+    },
+    {
+      "epoch": 4.180645161290323,
+      "grad_norm": 0.20078836381435394,
+      "learning_rate": 2.9263101785268253e-07,
+      "loss": 0.0012,
+      "num_input_tokens_seen": 1107520,
+      "step": 162
+    },
+    {
+      "epoch": 4.2064516129032254,
+      "grad_norm": 0.00893727969378233,
+      "learning_rate": 2.7248368952908055e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1114208,
+      "step": 163
+    },
+    {
+      "epoch": 4.232258064516129,
+      "grad_norm": 0.027001328766345978,
+      "learning_rate": 2.53014884252083e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1120912,
+      "step": 164
+    },
+    {
+      "epoch": 4.258064516129032,
+      "grad_norm": 0.01514577865600586,
+      "learning_rate": 2.3423053240837518e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1127872,
+      "step": 165
+    },
+    {
+      "epoch": 4.283870967741936,
+      "grad_norm": 0.0690593495965004,
+      "learning_rate": 2.1613635589349756e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1134352,
+      "step": 166
+    },
+    {
+      "epoch": 4.309677419354839,
+      "grad_norm": 0.021051278337836266,
+      "learning_rate": 1.9873786636889908e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1141440,
+      "step": 167
+    },
+    {
+      "epoch": 4.335483870967742,
+      "grad_norm": 0.3527485132217407,
+      "learning_rate": 1.8204036358303173e-07,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 1148096,
+      "step": 168
+    },
+    {
+      "epoch": 4.361290322580645,
+      "grad_norm": 0.09119685739278793,
+      "learning_rate": 1.6604893375699594e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1154672,
+      "step": 169
+    },
+    {
+      "epoch": 4.387096774193548,
+      "grad_norm": 0.014869263395667076,
+      "learning_rate": 1.507684480352292e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1161744,
+      "step": 170
+    },
+    {
+      "epoch": 4.412903225806452,
+      "grad_norm": 0.16110821068286896,
+      "learning_rate": 1.362035610017079e-07,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 1168944,
+      "step": 171
+    },
+    {
+      "epoch": 4.438709677419355,
+      "grad_norm": 0.05959083512425423,
+      "learning_rate": 1.223587092621162e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1175792,
+      "step": 172
+    },
+    {
+      "epoch": 4.464516129032258,
+      "grad_norm": 0.15646237134933472,
+      "learning_rate": 1.0923811009241142e-07,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1182800,
+      "step": 173
+    },
+    {
+      "epoch": 4.490322580645161,
+      "grad_norm": 0.024713285267353058,
+      "learning_rate": 9.684576015420277e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1189568,
+      "step": 174
+    },
+    {
+      "epoch": 4.516129032258064,
+      "grad_norm": 0.0995955690741539,
+      "learning_rate": 8.518543427732951e-08,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1196480,
+      "step": 175
+    },
+    {
+      "epoch": 4.541935483870968,
+      "grad_norm": 0.5925021767616272,
+      "learning_rate": 7.426068431000883e-08,
+      "loss": 0.0076,
+      "num_input_tokens_seen": 1203728,
+      "step": 176
+    },
+    {
+      "epoch": 4.567741935483871,
+      "grad_norm": 0.07701185345649719,
+      "learning_rate": 6.407483803691216e-08,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1210320,
+      "step": 177
+    },
+    {
+      "epoch": 4.593548387096774,
+      "grad_norm": 0.5693293213844299,
+      "learning_rate": 5.463099816548578e-08,
+      "loss": 0.004,
+      "num_input_tokens_seen": 1217248,
+      "step": 178
+    },
+    {
+      "epoch": 4.619354838709677,
+      "grad_norm": 0.025556327775120735,
+      "learning_rate": 4.593204138084006e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1224192,
+      "step": 179
+    },
+    {
+      "epoch": 4.645161290322581,
+      "grad_norm": 0.14782366156578064,
+      "learning_rate": 3.798061746947995e-08,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1230896,
+      "step": 180
+    },
+    {
+      "epoch": 4.670967741935484,
+      "grad_norm": 0.06809692084789276,
+      "learning_rate": 3.077914851215585e-08,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1237840,
+      "step": 181
+    },
+    {
+      "epoch": 4.6967741935483875,
+      "grad_norm": 0.01677132397890091,
+      "learning_rate": 2.4329828146074096e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1244688,
+      "step": 182
+    },
+    {
+      "epoch": 4.72258064516129,
+      "grad_norm": 0.019206374883651733,
+      "learning_rate": 1.8634620896695044e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1251584,
+      "step": 183
+    },
+    {
+      "epoch": 4.748387096774193,
+      "grad_norm": 0.02283065766096115,
+      "learning_rate": 1.3695261579316776e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1258448,
+      "step": 184
+    },
+    {
+      "epoch": 4.774193548387097,
+      "grad_norm": 1.8589617013931274,
+      "learning_rate": 9.513254770636138e-09,
+      "loss": 0.0167,
+      "num_input_tokens_seen": 1265440,
+      "step": 185
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 0.03213285654783249,
+      "learning_rate": 6.089874350439507e-09,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1272080,
+      "step": 186
+    },
+    {
+      "epoch": 4.825806451612904,
+      "grad_norm": 0.11905555427074432,
+      "learning_rate": 3.4261631135654174e-09,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1278976,
+      "step": 187
+    },
+    {
+      "epoch": 4.851612903225806,
+      "grad_norm": 0.384997695684433,
+      "learning_rate": 1.5229324522605949e-09,
+      "loss": 0.0024,
+      "num_input_tokens_seen": 1285760,
+      "step": 188
+    },
+    {
+      "epoch": 4.877419354838709,
+      "grad_norm": 0.08503298461437225,
+      "learning_rate": 3.8076210902182607e-10,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1292368,
+      "step": 189
+    },
+    {
+      "epoch": 4.903225806451613,
+      "grad_norm": 0.4136364758014679,
+      "learning_rate": 0.0,
+      "loss": 0.0017,
+      "num_input_tokens_seen": 1299120,
+      "step": 190
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 190,
+  "num_input_tokens_seen": 1299120,
+  "num_train_epochs": 5,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.150239379108659e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-190/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fea12da3624a50d04db32839a9390a4c8041b3756454a76a3011387f179a4228
+size 6584

checkpoint-190/zero_to_fp32.py

@@ -0,0 +1,604 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                                          exclude_frozen_parameters)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states,
+                                               exclude_frozen_parameters):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    if not exclude_frozen_parameters:
+        _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir, exclude_frozen_parameters)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None, exclude_frozen_parameters=False):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+        - ``exclude_frozen_parameters``: exclude frozen parameters
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag, exclude_frozen_parameters)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("--exclude_frozen_parameters", action='store_true', help="exclude frozen parameters")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir,
+                                               args.output_file,
+                                               tag=args.tag,
+                                               exclude_frozen_parameters=args.exclude_frozen_parameters)

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.42.3",
+  "use_cache": false,
+  "vocab_size": 32000
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.9,
+  "transformers_version": "4.42.3"
+}

llamaboard_config.yaml

@@ -0,0 +1,65 @@
+top.booster: auto
+top.checkpoint_path: null
+top.finetuning_type: full
+top.model_name: LLaMA2-7B-Chat
+top.quantization_bit: none
+top.quantization_method: bitsandbytes
+top.rope_scaling: none
+top.template: llama2
+top.visual_inputs: false
+train.additional_target: ''
+train.badam_mode: layer
+train.badam_switch_interval: 50
+train.badam_switch_mode: ascending
+train.badam_update_ratio: 0.05
+train.batch_size: 2
+train.compute_type: bf16
+train.create_new_adapter: false
+train.cutoff_len: 1024
+train.dataset:
+- truth_train_0716_2
+train.dataset_dir: data
+train.ds_offload: false
+train.ds_stage: '2'
+train.freeze_extra_modules: ''
+train.freeze_trainable_layers: 2
+train.freeze_trainable_modules: all
+train.galore_rank: 16
+train.galore_scale: 0.25
+train.galore_target: all
+train.galore_update_interval: 200
+train.gradient_accumulation_steps: 8
+train.learning_rate: 5e-6
+train.logging_steps: 1
+train.lora_alpha: 16
+train.lora_dropout: 0
+train.lora_rank: 8
+train.lora_target: ''
+train.loraplus_lr_ratio: 0
+train.lr_scheduler_type: cosine
+train.max_grad_norm: '1.0'
+train.max_samples: '100000'
+train.neat_packing: false
+train.neftune_alpha: 0
+train.num_train_epochs: '5.0'
+train.optim: adamw_torch
+train.packing: false
+train.ppo_score_norm: false
+train.ppo_whiten_rewards: false
+train.pref_beta: 0.1
+train.pref_ftx: 0
+train.pref_loss: sigmoid
+train.report_to: false
+train.resize_vocab: false
+train.reward_model: null
+train.save_steps: 1000
+train.shift_attn: false
+train.training_stage: Supervised Fine-Tuning
+train.use_badam: false
+train.use_dora: false
+train.use_galore: false
+train.use_llama_pro: false
+train.use_pissa: false
+train.use_rslora: false
+train.val_size: 0
+train.warmup_steps: 10

model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af6cb31a20c84677f5085cfa5e077f8442dc10afc0d91be75e4e150668712487
+size 4938985352

model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03c38c9aa8e45f6b755a47ad5ba015f1b7c2daf7c1459bdb1b15cb876344adc0
+size 4947390880

model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7175530932f0eafafa2a482d31b7f8df84aa9dd5f3b5e23dd168cdeb053363a7
+size 3590488816

model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "total_size": 13476831232
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00003-of-00003.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00002-of-00003.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00003-of-00003.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00001-of-00003.safetensors",
+    "model.norm.weight": "model-00003-of-00003.safetensors"
+  }
+}

running_log.txt

@@ -0,0 +1,633 @@
+07/16/2024 16:50:17 - INFO - llamafactory.hparams.parser - Process rank: 7, device: cuda:7, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+[INFO|parser.py:325] 2024-07-16 16:50:17,965 >> Process rank: 0, device: cuda:0, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+07/16/2024 16:50:18 - INFO - llamafactory.hparams.parser - Process rank: 4, device: cuda:4, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+07/16/2024 16:50:18 - INFO - llamafactory.hparams.parser - Process rank: 6, device: cuda:6, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+07/16/2024 16:50:18 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 16:50:18 - INFO - llamafactory.hparams.parser - Process rank: 5, device: cuda:5, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+07/16/2024 16:50:18 - INFO - llamafactory.hparams.parser - Process rank: 1, device: cuda:1, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 16:50:18,169 >> loading file tokenizer.model from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/tokenizer.model
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 16:50:18,169 >> loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/tokenizer.json
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 16:50:18,170 >> loading file added_tokens.json from cache at None
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 16:50:18,170 >> loading file special_tokens_map.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/special_tokens_map.json
+
+[INFO|tokenization_utils_base.py:2161] 2024-07-16 16:50:18,170 >> loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/tokenizer_config.json
+
+07/16/2024 16:50:18 - INFO - llamafactory.data.template - Add pad token: </s>
+
+[INFO|template.py:372] 2024-07-16 16:50:18,281 >> Add pad token: </s>
+
+[INFO|loader.py:50] 2024-07-16 16:50:18,282 >> Loading dataset 0716_truthfulqa_benchmark_train_2.json...
+
+07/16/2024 16:50:18 - INFO - llamafactory.hparams.parser - Process rank: 2, device: cuda:2, n_gpu: 1, distributed training: True, compute dtype: torch.bfloat16
+
+07/16/2024 16:50:18 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 16:50:18 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 16:50:18 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 16:50:18 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 16:50:18 - INFO - llamafactory.data.template - Add pad token: </s>
+
+07/16/2024 16:50:19 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train_2.json...
+
+07/16/2024 16:50:19 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train_2.json...
+
+07/16/2024 16:50:19 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train_2.json...
+
+07/16/2024 16:50:19 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train_2.json...
+
+07/16/2024 16:50:19 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train_2.json...
+
+07/16/2024 16:50:19 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train_2.json...
+
+07/16/2024 16:50:19 - INFO - llamafactory.data.loader - Loading dataset 0716_truthfulqa_benchmark_train_2.json...
+
+[INFO|configuration_utils.py:733] 2024-07-16 16:50:20,327 >> loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/config.json
+
+[INFO|configuration_utils.py:800] 2024-07-16 16:50:20,328 >> Model config LlamaConfig {
+  "_name_or_path": "meta-llama/Llama-2-7b-chat-hf",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.42.3",
+  "use_cache": true,
+  "vocab_size": 32000
+}
+
+
+[INFO|modeling_utils.py:3556] 2024-07-16 16:50:20,350 >> loading weights file model.safetensors from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/model.safetensors.index.json
+
+[INFO|modeling_utils.py:1531] 2024-07-16 16:50:20,351 >> Instantiating LlamaForCausalLM model under default dtype torch.bfloat16.
+
+[INFO|configuration_utils.py:1000] 2024-07-16 16:50:20,352 >> Generate config GenerationConfig {
+  "bos_token_id": 1,
+  "eos_token_id": 2
+}
+
+
+[INFO|modeling_utils.py:4364] 2024-07-16 16:50:37,558 >> All model checkpoint weights were used when initializing LlamaForCausalLM.
+
+
+[INFO|modeling_utils.py:4372] 2024-07-16 16:50:37,559 >> All the weights of LlamaForCausalLM were initialized from the model checkpoint at meta-llama/Llama-2-7b-chat-hf.
+If your task is similar to the task the model of the checkpoint was trained on, you can already use LlamaForCausalLM for predictions without further training.
+
+[INFO|configuration_utils.py:955] 2024-07-16 16:50:37,738 >> loading configuration file generation_config.json from cache at /root/.cache/huggingface/hub/models--meta-llama--Llama-2-7b-chat-hf/snapshots/f5db02db724555f92da89c216ac04704f23d4590/generation_config.json
+
+[INFO|configuration_utils.py:1000] 2024-07-16 16:50:37,738 >> Generate config GenerationConfig {
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 2,
+  "max_length": 4096,
+  "pad_token_id": 0,
+  "temperature": 0.6,
+  "top_p": 0.9
+}
+
+
+[INFO|checkpointing.py:103] 2024-07-16 16:50:37,746 >> Gradient checkpointing enabled.
+
+[INFO|attention.py:80] 2024-07-16 16:50:37,746 >> Using torch SDPA for faster training and inference.
+
+[INFO|adapter.py:302] 2024-07-16 16:50:37,746 >> Upcasting trainable params to float32.
+
+[INFO|adapter.py:48] 2024-07-16 16:50:37,746 >> Fine-tuning method: Full
+
+[INFO|loader.py:196] 2024-07-16 16:50:37,798 >> trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 16:50:37 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 16:50:37 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 16:50:37 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 16:50:37 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 16:50:37 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+[INFO|trainer.py:642] 2024-07-16 16:50:37,804 >> Using auto half precision backend
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.checkpointing - Gradient checkpointing enabled.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.model_utils.attention - Using torch SDPA for faster training and inference.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.adapter - Fine-tuning method: Full
+
+07/16/2024 16:50:38 - INFO - llamafactory.model.loader - trainable params: 6,738,415,616 || all params: 6,738,415,616 || trainable%: 100.0000
+
+[INFO|trainer.py:2128] 2024-07-16 16:50:57,380 >> ***** Running training *****
+
+[INFO|trainer.py:2129] 2024-07-16 16:50:57,380 >>   Num examples = 4,958
+
+[INFO|trainer.py:2130] 2024-07-16 16:50:57,380 >>   Num Epochs = 5
+
+[INFO|trainer.py:2131] 2024-07-16 16:50:57,380 >>   Instantaneous batch size per device = 2
+
+[INFO|trainer.py:2134] 2024-07-16 16:50:57,380 >>   Total train batch size (w. parallel, distributed & accumulation) = 128
+
+[INFO|trainer.py:2135] 2024-07-16 16:50:57,380 >>   Gradient Accumulation steps = 8
+
+[INFO|trainer.py:2136] 2024-07-16 16:50:57,380 >>   Total optimization steps = 190
+
+[INFO|trainer.py:2137] 2024-07-16 16:50:57,381 >>   Number of trainable parameters = 6,738,415,616
+
+[INFO|callbacks.py:310] 2024-07-16 16:51:09,938 >> {'loss': 8.2514, 'learning_rate': 5.0000e-07, 'epoch': 0.03, 'throughput': 545.43}
+
+[INFO|callbacks.py:310] 2024-07-16 16:51:20,951 >> {'loss': 8.2793, 'learning_rate': 1.0000e-06, 'epoch': 0.05, 'throughput': 584.51}
+
+[INFO|callbacks.py:310] 2024-07-16 16:51:31,943 >> {'loss': 8.1700, 'learning_rate': 1.5000e-06, 'epoch': 0.08, 'throughput': 598.15}
+
+[INFO|callbacks.py:310] 2024-07-16 16:51:42,923 >> {'loss': 7.6197, 'learning_rate': 2.0000e-06, 'epoch': 0.10, 'throughput': 609.21}
+
+[INFO|callbacks.py:310] 2024-07-16 16:51:53,920 >> {'loss': 6.9491, 'learning_rate': 2.5000e-06, 'epoch': 0.13, 'throughput': 612.41}
+
+[INFO|callbacks.py:310] 2024-07-16 16:52:04,919 >> {'loss': 5.2054, 'learning_rate': 3.0000e-06, 'epoch': 0.15, 'throughput': 613.36}
+
+[INFO|callbacks.py:310] 2024-07-16 16:52:15,920 >> {'loss': 4.8642, 'learning_rate': 3.5000e-06, 'epoch': 0.18, 'throughput': 615.05}
+
+[INFO|callbacks.py:310] 2024-07-16 16:52:26,924 >> {'loss': 3.2874, 'learning_rate': 4.0000e-06, 'epoch': 0.21, 'throughput': 615.94}
+
+[INFO|callbacks.py:310] 2024-07-16 16:52:37,962 >> {'loss': 2.6310, 'learning_rate': 4.5000e-06, 'epoch': 0.23, 'throughput': 613.25}
+
+[INFO|callbacks.py:310] 2024-07-16 16:52:48,988 >> {'loss': 0.6982, 'learning_rate': 5.0000e-06, 'epoch': 0.26, 'throughput': 613.59}
+
+[INFO|callbacks.py:310] 2024-07-16 16:53:00,018 >> {'loss': 0.3276, 'learning_rate': 4.9996e-06, 'epoch': 0.28, 'throughput': 613.98}
+
+[INFO|callbacks.py:310] 2024-07-16 16:53:11,029 >> {'loss': 0.2930, 'learning_rate': 4.9985e-06, 'epoch': 0.31, 'throughput': 615.72}
+
+[INFO|callbacks.py:310] 2024-07-16 16:53:22,020 >> {'loss': 0.2129, 'learning_rate': 4.9966e-06, 'epoch': 0.34, 'throughput': 615.94}
+
+[INFO|callbacks.py:310] 2024-07-16 16:53:33,021 >> {'loss': 0.4712, 'learning_rate': 4.9939e-06, 'epoch': 0.36, 'throughput': 616.20}
+
+[INFO|callbacks.py:310] 2024-07-16 16:53:44,003 >> {'loss': 0.2350, 'learning_rate': 4.9905e-06, 'epoch': 0.39, 'throughput': 617.55}
+
+[INFO|callbacks.py:310] 2024-07-16 16:53:55,026 >> {'loss': 0.2020, 'learning_rate': 4.9863e-06, 'epoch': 0.41, 'throughput': 618.05}
+
+[INFO|callbacks.py:310] 2024-07-16 16:54:06,023 >> {'loss': 0.1981, 'learning_rate': 4.9814e-06, 'epoch': 0.44, 'throughput': 617.73}
+
+[INFO|callbacks.py:310] 2024-07-16 16:54:17,028 >> {'loss': 0.1517, 'learning_rate': 4.9757e-06, 'epoch': 0.46, 'throughput': 617.10}
+
+[INFO|callbacks.py:310] 2024-07-16 16:54:28,037 >> {'loss': 0.4335, 'learning_rate': 4.9692e-06, 'epoch': 0.49, 'throughput': 617.28}
+
+[INFO|callbacks.py:310] 2024-07-16 16:54:39,050 >> {'loss': 0.3609, 'learning_rate': 4.9620e-06, 'epoch': 0.52, 'throughput': 617.29}
+
+[INFO|callbacks.py:310] 2024-07-16 16:54:50,034 >> {'loss': 0.1708, 'learning_rate': 4.9541e-06, 'epoch': 0.54, 'throughput': 618.54}
+
+[INFO|callbacks.py:310] 2024-07-16 16:55:01,020 >> {'loss': 0.2277, 'learning_rate': 4.9454e-06, 'epoch': 0.57, 'throughput': 617.70}
+
+[INFO|callbacks.py:310] 2024-07-16 16:55:12,039 >> {'loss': 0.3437, 'learning_rate': 4.9359e-06, 'epoch': 0.59, 'throughput': 617.93}
+
+[INFO|callbacks.py:310] 2024-07-16 16:55:23,067 >> {'loss': 0.2229, 'learning_rate': 4.9257e-06, 'epoch': 0.62, 'throughput': 619.02}
+
+[INFO|callbacks.py:310] 2024-07-16 16:55:34,096 >> {'loss': 0.1242, 'learning_rate': 4.9148e-06, 'epoch': 0.65, 'throughput': 617.82}
+
+[INFO|callbacks.py:310] 2024-07-16 16:55:45,128 >> {'loss': 0.2117, 'learning_rate': 4.9032e-06, 'epoch': 0.67, 'throughput': 617.94}
+
+[INFO|callbacks.py:310] 2024-07-16 16:55:56,152 >> {'loss': 0.2706, 'learning_rate': 4.8908e-06, 'epoch': 0.70, 'throughput': 618.70}
+
+[INFO|callbacks.py:310] 2024-07-16 16:56:07,175 >> {'loss': 0.2084, 'learning_rate': 4.8776e-06, 'epoch': 0.72, 'throughput': 618.27}
+
+[INFO|callbacks.py:310] 2024-07-16 16:56:18,165 >> {'loss': 0.0981, 'learning_rate': 4.8638e-06, 'epoch': 0.75, 'throughput': 618.39}
+
+[INFO|callbacks.py:310] 2024-07-16 16:56:29,154 >> {'loss': 0.1600, 'learning_rate': 4.8492e-06, 'epoch': 0.77, 'throughput': 618.50}
+
+[INFO|callbacks.py:310] 2024-07-16 16:56:40,149 >> {'loss': 0.1614, 'learning_rate': 4.8340e-06, 'epoch': 0.80, 'throughput': 617.80}
+
+[INFO|callbacks.py:310] 2024-07-16 16:56:51,163 >> {'loss': 0.1742, 'learning_rate': 4.8180e-06, 'epoch': 0.83, 'throughput': 617.47}
+
+[INFO|callbacks.py:310] 2024-07-16 16:57:02,179 >> {'loss': 0.1107, 'learning_rate': 4.8013e-06, 'epoch': 0.85, 'throughput': 617.90}
+
+[INFO|callbacks.py:310] 2024-07-16 16:57:13,192 >> {'loss': 0.0822, 'learning_rate': 4.7839e-06, 'epoch': 0.88, 'throughput': 617.42}
+
+[INFO|callbacks.py:310] 2024-07-16 16:57:24,203 >> {'loss': 0.1873, 'learning_rate': 4.7658e-06, 'epoch': 0.90, 'throughput': 617.01}
+
+[INFO|callbacks.py:310] 2024-07-16 16:57:35,243 >> {'loss': 0.2375, 'learning_rate': 4.7470e-06, 'epoch': 0.93, 'throughput': 616.94}
+
+[INFO|callbacks.py:310] 2024-07-16 16:57:46,259 >> {'loss': 0.2667, 'learning_rate': 4.7275e-06, 'epoch': 0.95, 'throughput': 617.73}
+
+[INFO|callbacks.py:310] 2024-07-16 16:57:57,247 >> {'loss': 0.1547, 'learning_rate': 4.7074e-06, 'epoch': 0.98, 'throughput': 618.14}
+
+[INFO|callbacks.py:310] 2024-07-16 16:58:08,231 >> {'loss': 0.1662, 'learning_rate': 4.6865e-06, 'epoch': 1.01, 'throughput': 618.69}
+
+[INFO|callbacks.py:310] 2024-07-16 16:58:19,239 >> {'loss': 0.0808, 'learning_rate': 4.6651e-06, 'epoch': 1.03, 'throughput': 618.41}
+
+[INFO|callbacks.py:310] 2024-07-16 16:58:30,245 >> {'loss': 0.0884, 'learning_rate': 4.6429e-06, 'epoch': 1.06, 'throughput': 618.04}
+
+[INFO|callbacks.py:310] 2024-07-16 16:58:41,281 >> {'loss': 0.0883, 'learning_rate': 4.6201e-06, 'epoch': 1.08, 'throughput': 618.55}
+
+[INFO|callbacks.py:310] 2024-07-16 16:58:52,323 >> {'loss': 0.0562, 'learning_rate': 4.5967e-06, 'epoch': 1.11, 'throughput': 618.59}
+
+[INFO|callbacks.py:310] 2024-07-16 16:59:03,347 >> {'loss': 0.0856, 'learning_rate': 4.5726e-06, 'epoch': 1.14, 'throughput': 618.38}
+
+[INFO|callbacks.py:310] 2024-07-16 16:59:14,361 >> {'loss': 0.0612, 'learning_rate': 4.5479e-06, 'epoch': 1.16, 'throughput': 618.26}
+
+[INFO|callbacks.py:310] 2024-07-16 16:59:25,365 >> {'loss': 0.0944, 'learning_rate': 4.5225e-06, 'epoch': 1.19, 'throughput': 618.35}
+
+[INFO|callbacks.py:310] 2024-07-16 16:59:36,390 >> {'loss': 0.0624, 'learning_rate': 4.4966e-06, 'epoch': 1.21, 'throughput': 618.23}
+
+[INFO|callbacks.py:310] 2024-07-16 16:59:47,376 >> {'loss': 0.0363, 'learning_rate': 4.4700e-06, 'epoch': 1.24, 'throughput': 618.21}
+
+[INFO|callbacks.py:310] 2024-07-16 16:59:58,399 >> {'loss': 0.1039, 'learning_rate': 4.4429e-06, 'epoch': 1.26, 'throughput': 618.16}
+
+[INFO|callbacks.py:310] 2024-07-16 17:00:09,394 >> {'loss': 0.0488, 'learning_rate': 4.4151e-06, 'epoch': 1.29, 'throughput': 618.19}
+
+[INFO|callbacks.py:310] 2024-07-16 17:00:20,399 >> {'loss': 0.0613, 'learning_rate': 4.3868e-06, 'epoch': 1.32, 'throughput': 618.44}
+
+[INFO|callbacks.py:310] 2024-07-16 17:00:31,407 >> {'loss': 0.0700, 'learning_rate': 4.3579e-06, 'epoch': 1.34, 'throughput': 618.12}
+
+[INFO|callbacks.py:310] 2024-07-16 17:00:42,424 >> {'loss': 0.0463, 'learning_rate': 4.3284e-06, 'epoch': 1.37, 'throughput': 618.08}
+
+[INFO|callbacks.py:310] 2024-07-16 17:00:53,462 >> {'loss': 0.0671, 'learning_rate': 4.2983e-06, 'epoch': 1.39, 'throughput': 618.15}
+
+[INFO|callbacks.py:310] 2024-07-16 17:01:04,468 >> {'loss': 0.0428, 'learning_rate': 4.2678e-06, 'epoch': 1.42, 'throughput': 618.43}
+
+[INFO|callbacks.py:310] 2024-07-16 17:01:15,463 >> {'loss': 0.0678, 'learning_rate': 4.2366e-06, 'epoch': 1.45, 'throughput': 618.43}
+
+[INFO|callbacks.py:310] 2024-07-16 17:01:26,456 >> {'loss': 0.0476, 'learning_rate': 4.2050e-06, 'epoch': 1.47, 'throughput': 618.38}
+
+[INFO|callbacks.py:310] 2024-07-16 17:01:37,444 >> {'loss': 0.0442, 'learning_rate': 4.1728e-06, 'epoch': 1.50, 'throughput': 618.82}
+
+[INFO|callbacks.py:310] 2024-07-16 17:01:48,428 >> {'loss': 0.0336, 'learning_rate': 4.1401e-06, 'epoch': 1.52, 'throughput': 619.09}
+
+[INFO|callbacks.py:310] 2024-07-16 17:01:59,445 >> {'loss': 0.0460, 'learning_rate': 4.1070e-06, 'epoch': 1.55, 'throughput': 618.77}
+
+[INFO|callbacks.py:310] 2024-07-16 17:02:10,459 >> {'loss': 0.0416, 'learning_rate': 4.0733e-06, 'epoch': 1.57, 'throughput': 618.46}
+
+[INFO|callbacks.py:310] 2024-07-16 17:02:21,470 >> {'loss': 0.0649, 'learning_rate': 4.0392e-06, 'epoch': 1.60, 'throughput': 618.87}
+
+[INFO|callbacks.py:310] 2024-07-16 17:02:32,483 >> {'loss': 0.0591, 'learning_rate': 4.0045e-06, 'epoch': 1.63, 'throughput': 619.05}
+
+[INFO|callbacks.py:310] 2024-07-16 17:02:43,490 >> {'loss': 0.0318, 'learning_rate': 3.9695e-06, 'epoch': 1.65, 'throughput': 618.83}
+
+[INFO|callbacks.py:310] 2024-07-16 17:02:54,478 >> {'loss': 0.0462, 'learning_rate': 3.9339e-06, 'epoch': 1.68, 'throughput': 618.87}
+
+[INFO|callbacks.py:310] 2024-07-16 17:03:05,466 >> {'loss': 0.0465, 'learning_rate': 3.8980e-06, 'epoch': 1.70, 'throughput': 618.98}
+
+[INFO|callbacks.py:310] 2024-07-16 17:03:16,480 >> {'loss': 0.0316, 'learning_rate': 3.8616e-06, 'epoch': 1.73, 'throughput': 619.15}
+
+[INFO|callbacks.py:310] 2024-07-16 17:03:27,480 >> {'loss': 0.1000, 'learning_rate': 3.8248e-06, 'epoch': 1.75, 'throughput': 619.38}
+
+[INFO|callbacks.py:310] 2024-07-16 17:03:38,513 >> {'loss': 0.0711, 'learning_rate': 3.7876e-06, 'epoch': 1.78, 'throughput': 619.25}
+
+[INFO|callbacks.py:310] 2024-07-16 17:03:49,506 >> {'loss': 0.0494, 'learning_rate': 3.7500e-06, 'epoch': 1.81, 'throughput': 619.69}
+
+[INFO|callbacks.py:310] 2024-07-16 17:04:00,519 >> {'loss': 0.0618, 'learning_rate': 3.7120e-06, 'epoch': 1.83, 'throughput': 619.64}
+
+[INFO|callbacks.py:310] 2024-07-16 17:04:11,528 >> {'loss': 0.0511, 'learning_rate': 3.6737e-06, 'epoch': 1.86, 'throughput': 619.61}
+
+[INFO|callbacks.py:310] 2024-07-16 17:04:22,525 >> {'loss': 0.0464, 'learning_rate': 3.6350e-06, 'epoch': 1.88, 'throughput': 619.64}
+
+[INFO|callbacks.py:310] 2024-07-16 17:04:33,509 >> {'loss': 0.0331, 'learning_rate': 3.5959e-06, 'epoch': 1.91, 'throughput': 620.02}
+
+[INFO|callbacks.py:310] 2024-07-16 17:04:44,504 >> {'loss': 0.0706, 'learning_rate': 3.5565e-06, 'epoch': 1.94, 'throughput': 620.04}
+
+[INFO|callbacks.py:310] 2024-07-16 17:04:55,490 >> {'loss': 0.0442, 'learning_rate': 3.5168e-06, 'epoch': 1.96, 'throughput': 620.14}
+
+[INFO|callbacks.py:310] 2024-07-16 17:05:06,484 >> {'loss': 0.0420, 'learning_rate': 3.4768e-06, 'epoch': 1.99, 'throughput': 619.89}
+
+[INFO|callbacks.py:310] 2024-07-16 17:05:17,496 >> {'loss': 0.0210, 'learning_rate': 3.4365e-06, 'epoch': 2.01, 'throughput': 619.84}
+
+[INFO|callbacks.py:310] 2024-07-16 17:05:28,503 >> {'loss': 0.0094, 'learning_rate': 3.3959e-06, 'epoch': 2.04, 'throughput': 619.91}
+
+[INFO|callbacks.py:310] 2024-07-16 17:05:39,520 >> {'loss': 0.0021, 'learning_rate': 3.3551e-06, 'epoch': 2.06, 'throughput': 620.26}
+
+[INFO|callbacks.py:310] 2024-07-16 17:05:50,557 >> {'loss': 0.0146, 'learning_rate': 3.3139e-06, 'epoch': 2.09, 'throughput': 620.13}
+
+[INFO|callbacks.py:310] 2024-07-16 17:06:01,557 >> {'loss': 0.0237, 'learning_rate': 3.2725e-06, 'epoch': 2.12, 'throughput': 620.48}
+
+[INFO|callbacks.py:310] 2024-07-16 17:06:12,563 >> {'loss': 0.0031, 'learning_rate': 3.2309e-06, 'epoch': 2.14, 'throughput': 620.57}
+
+[INFO|callbacks.py:310] 2024-07-16 17:06:23,576 >> {'loss': 0.0034, 'learning_rate': 3.1891e-06, 'epoch': 2.17, 'throughput': 620.54}
+
+[INFO|callbacks.py:310] 2024-07-16 17:06:34,571 >> {'loss': 0.0045, 'learning_rate': 3.1470e-06, 'epoch': 2.19, 'throughput': 620.66}
+
+[INFO|callbacks.py:310] 2024-07-16 17:06:45,568 >> {'loss': 0.0031, 'learning_rate': 3.1048e-06, 'epoch': 2.22, 'throughput': 620.62}
+
+[INFO|callbacks.py:310] 2024-07-16 17:06:56,589 >> {'loss': 0.0341, 'learning_rate': 3.0624e-06, 'epoch': 2.25, 'throughput': 620.63}
+
+[INFO|callbacks.py:310] 2024-07-16 17:07:07,621 >> {'loss': 0.0095, 'learning_rate': 3.0198e-06, 'epoch': 2.27, 'throughput': 620.55}
+
+[INFO|callbacks.py:310] 2024-07-16 17:07:18,643 >> {'loss': 0.0459, 'learning_rate': 2.9770e-06, 'epoch': 2.30, 'throughput': 620.80}
+
+[INFO|callbacks.py:310] 2024-07-16 17:07:29,659 >> {'loss': 0.0104, 'learning_rate': 2.9341e-06, 'epoch': 2.32, 'throughput': 620.80}
+
+[INFO|callbacks.py:310] 2024-07-16 17:07:40,662 >> {'loss': 0.0201, 'learning_rate': 2.8911e-06, 'epoch': 2.35, 'throughput': 620.51}
+
+[INFO|callbacks.py:310] 2024-07-16 17:07:51,636 >> {'loss': 0.0021, 'learning_rate': 2.8479e-06, 'epoch': 2.37, 'throughput': 620.75}
+
+[INFO|callbacks.py:310] 2024-07-16 17:08:02,651 >> {'loss': 0.0430, 'learning_rate': 2.8047e-06, 'epoch': 2.40, 'throughput': 620.64}
+
+[INFO|callbacks.py:310] 2024-07-16 17:08:13,671 >> {'loss': 0.0207, 'learning_rate': 2.7613e-06, 'epoch': 2.43, 'throughput': 620.61}
+
+[INFO|callbacks.py:310] 2024-07-16 17:08:24,677 >> {'loss': 0.0148, 'learning_rate': 2.7179e-06, 'epoch': 2.45, 'throughput': 620.69}
+
+[INFO|callbacks.py:310] 2024-07-16 17:08:35,700 >> {'loss': 0.0040, 'learning_rate': 2.6744e-06, 'epoch': 2.48, 'throughput': 620.55}
+
+[INFO|callbacks.py:310] 2024-07-16 17:08:46,703 >> {'loss': 0.0131, 'learning_rate': 2.6308e-06, 'epoch': 2.50, 'throughput': 620.54}
+
+[INFO|callbacks.py:310] 2024-07-16 17:08:57,742 >> {'loss': 0.0455, 'learning_rate': 2.5872e-06, 'epoch': 2.53, 'throughput': 620.37}
+
+[INFO|callbacks.py:310] 2024-07-16 17:09:08,772 >> {'loss': 0.0031, 'learning_rate': 2.5436e-06, 'epoch': 2.55, 'throughput': 620.27}
+
+[INFO|callbacks.py:310] 2024-07-16 17:09:19,760 >> {'loss': 0.0099, 'learning_rate': 2.5000e-06, 'epoch': 2.58, 'throughput': 620.49}
+
+[INFO|callbacks.py:310] 2024-07-16 17:09:30,748 >> {'loss': 0.0797, 'learning_rate': 2.4564e-06, 'epoch': 2.61, 'throughput': 620.49}
+
+[INFO|callbacks.py:310] 2024-07-16 17:09:41,738 >> {'loss': 0.0059, 'learning_rate': 2.4128e-06, 'epoch': 2.63, 'throughput': 620.63}
+
+[INFO|callbacks.py:310] 2024-07-16 17:09:52,737 >> {'loss': 0.0438, 'learning_rate': 2.3692e-06, 'epoch': 2.66, 'throughput': 620.38}
+
+[INFO|callbacks.py:310] 2024-07-16 17:10:03,734 >> {'loss': 0.0149, 'learning_rate': 2.3256e-06, 'epoch': 2.68, 'throughput': 620.63}
+
+[INFO|callbacks.py:310] 2024-07-16 17:10:14,743 >> {'loss': 0.0126, 'learning_rate': 2.2821e-06, 'epoch': 2.71, 'throughput': 620.57}
+
+[INFO|callbacks.py:310] 2024-07-16 17:10:25,754 >> {'loss': 0.0255, 'learning_rate': 2.2387e-06, 'epoch': 2.74, 'throughput': 620.46}
+
+[INFO|callbacks.py:310] 2024-07-16 17:10:36,760 >> {'loss': 0.0048, 'learning_rate': 2.1953e-06, 'epoch': 2.76, 'throughput': 620.34}
+
+[INFO|callbacks.py:310] 2024-07-16 17:10:47,758 >> {'loss': 0.0142, 'learning_rate': 2.1521e-06, 'epoch': 2.79, 'throughput': 620.20}
+
+[INFO|callbacks.py:310] 2024-07-16 17:10:58,726 >> {'loss': 0.0193, 'learning_rate': 2.1089e-06, 'epoch': 2.81, 'throughput': 620.23}
+
+[INFO|callbacks.py:310] 2024-07-16 17:11:09,695 >> {'loss': 0.0055, 'learning_rate': 2.0659e-06, 'epoch': 2.84, 'throughput': 620.32}
+
+[INFO|callbacks.py:310] 2024-07-16 17:11:20,678 >> {'loss': 0.0144, 'learning_rate': 2.0230e-06, 'epoch': 2.86, 'throughput': 620.23}
+
+[INFO|callbacks.py:310] 2024-07-16 17:11:31,656 >> {'loss': 0.0272, 'learning_rate': 1.9802e-06, 'epoch': 2.89, 'throughput': 620.22}
+
+[INFO|callbacks.py:310] 2024-07-16 17:11:42,652 >> {'loss': 0.0101, 'learning_rate': 1.9376e-06, 'epoch': 2.92, 'throughput': 620.18}
+
+[INFO|callbacks.py:310] 2024-07-16 17:11:53,647 >> {'loss': 0.0109, 'learning_rate': 1.8952e-06, 'epoch': 2.94, 'throughput': 620.51}
+
+[INFO|callbacks.py:310] 2024-07-16 17:12:04,642 >> {'loss': 0.0180, 'learning_rate': 1.8530e-06, 'epoch': 2.97, 'throughput': 620.55}
+
+[INFO|callbacks.py:310] 2024-07-16 17:12:15,639 >> {'loss': 0.0141, 'learning_rate': 1.8109e-06, 'epoch': 2.99, 'throughput': 620.38}
+
+[INFO|callbacks.py:310] 2024-07-16 17:12:26,637 >> {'loss': 0.0057, 'learning_rate': 1.7691e-06, 'epoch': 3.02, 'throughput': 620.36}
+
+[INFO|callbacks.py:310] 2024-07-16 17:12:37,628 >> {'loss': 0.0063, 'learning_rate': 1.7275e-06, 'epoch': 3.05, 'throughput': 620.33}
+
+[INFO|callbacks.py:310] 2024-07-16 17:12:48,606 >> {'loss': 0.0138, 'learning_rate': 1.6861e-06, 'epoch': 3.07, 'throughput': 620.27}
+
+[INFO|callbacks.py:310] 2024-07-16 17:12:59,593 >> {'loss': 0.0011, 'learning_rate': 1.6449e-06, 'epoch': 3.10, 'throughput': 619.97}
+
+[INFO|callbacks.py:310] 2024-07-16 17:13:10,574 >> {'loss': 0.0006, 'learning_rate': 1.6041e-06, 'epoch': 3.12, 'throughput': 619.94}
+
+[INFO|callbacks.py:310] 2024-07-16 17:13:21,574 >> {'loss': 0.0055, 'learning_rate': 1.5635e-06, 'epoch': 3.15, 'throughput': 619.97}
+
+[INFO|callbacks.py:310] 2024-07-16 17:13:32,565 >> {'loss': 0.0011, 'learning_rate': 1.5232e-06, 'epoch': 3.17, 'throughput': 619.90}
+
+[INFO|callbacks.py:310] 2024-07-16 17:13:43,569 >> {'loss': 0.0173, 'learning_rate': 1.4832e-06, 'epoch': 3.20, 'throughput': 620.05}
+
+[INFO|callbacks.py:310] 2024-07-16 17:13:54,579 >> {'loss': 0.0027, 'learning_rate': 1.4435e-06, 'epoch': 3.23, 'throughput': 619.89}
+
+[INFO|callbacks.py:310] 2024-07-16 17:14:05,559 >> {'loss': 0.0029, 'learning_rate': 1.4041e-06, 'epoch': 3.25, 'throughput': 619.81}
+
+[INFO|callbacks.py:310] 2024-07-16 17:14:16,541 >> {'loss': 0.0003, 'learning_rate': 1.3650e-06, 'epoch': 3.28, 'throughput': 619.97}
+
+[INFO|callbacks.py:310] 2024-07-16 17:14:27,516 >> {'loss': 0.0007, 'learning_rate': 1.3263e-06, 'epoch': 3.30, 'throughput': 619.98}
+
+[INFO|callbacks.py:310] 2024-07-16 17:14:38,496 >> {'loss': 0.0080, 'learning_rate': 1.2880e-06, 'epoch': 3.33, 'throughput': 620.05}
+
+[INFO|callbacks.py:310] 2024-07-16 17:14:49,489 >> {'loss': 0.0004, 'learning_rate': 1.2500e-06, 'epoch': 3.35, 'throughput': 620.16}
+
+[INFO|callbacks.py:310] 2024-07-16 17:15:00,489 >> {'loss': 0.0049, 'learning_rate': 1.2124e-06, 'epoch': 3.38, 'throughput': 620.39}
+
+[INFO|callbacks.py:310] 2024-07-16 17:15:11,487 >> {'loss': 0.0012, 'learning_rate': 1.1752e-06, 'epoch': 3.41, 'throughput': 620.30}
+
+[INFO|callbacks.py:310] 2024-07-16 17:15:22,486 >> {'loss': 0.0044, 'learning_rate': 1.1384e-06, 'epoch': 3.43, 'throughput': 620.50}
+
+[INFO|callbacks.py:310] 2024-07-16 17:15:33,486 >> {'loss': 0.0017, 'learning_rate': 1.1020e-06, 'epoch': 3.46, 'throughput': 620.57}
+
+[INFO|callbacks.py:310] 2024-07-16 17:15:44,480 >> {'loss': 0.0003, 'learning_rate': 1.0661e-06, 'epoch': 3.48, 'throughput': 620.49}
+
+[INFO|callbacks.py:310] 2024-07-16 17:15:55,449 >> {'loss': 0.0099, 'learning_rate': 1.0305e-06, 'epoch': 3.51, 'throughput': 620.45}
+
+[INFO|callbacks.py:310] 2024-07-16 17:16:06,411 >> {'loss': 0.0068, 'learning_rate': 9.9546e-07, 'epoch': 3.54, 'throughput': 620.34}
+
+[INFO|callbacks.py:310] 2024-07-16 17:16:17,397 >> {'loss': 0.0025, 'learning_rate': 9.6085e-07, 'epoch': 3.56, 'throughput': 620.33}
+
+[INFO|callbacks.py:310] 2024-07-16 17:16:28,378 >> {'loss': 0.0004, 'learning_rate': 9.2670e-07, 'epoch': 3.59, 'throughput': 620.50}
+
+[INFO|callbacks.py:310] 2024-07-16 17:16:39,370 >> {'loss': 0.0101, 'learning_rate': 8.9303e-07, 'epoch': 3.61, 'throughput': 620.37}
+
+[INFO|callbacks.py:310] 2024-07-16 17:16:50,370 >> {'loss': 0.0068, 'learning_rate': 8.5985e-07, 'epoch': 3.64, 'throughput': 620.41}
+
+[INFO|callbacks.py:310] 2024-07-16 17:17:01,381 >> {'loss': 0.0007, 'learning_rate': 8.2717e-07, 'epoch': 3.66, 'throughput': 620.29}
+
+[INFO|callbacks.py:310] 2024-07-16 17:17:12,379 >> {'loss': 0.0161, 'learning_rate': 7.9500e-07, 'epoch': 3.69, 'throughput': 620.19}
+
+[INFO|callbacks.py:310] 2024-07-16 17:17:23,362 >> {'loss': 0.0115, 'learning_rate': 7.6335e-07, 'epoch': 3.72, 'throughput': 620.44}
+
+[INFO|callbacks.py:310] 2024-07-16 17:17:34,347 >> {'loss': 0.0052, 'learning_rate': 7.3223e-07, 'epoch': 3.74, 'throughput': 620.49}
+
+[INFO|callbacks.py:310] 2024-07-16 17:17:45,329 >> {'loss': 0.0098, 'learning_rate': 7.0165e-07, 'epoch': 3.77, 'throughput': 620.56}
+
+[INFO|callbacks.py:310] 2024-07-16 17:17:56,308 >> {'loss': 0.0005, 'learning_rate': 6.7162e-07, 'epoch': 3.79, 'throughput': 620.74}
+
+[INFO|callbacks.py:310] 2024-07-16 17:18:07,295 >> {'loss': 0.0012, 'learning_rate': 6.4214e-07, 'epoch': 3.82, 'throughput': 620.71}
+
+[INFO|callbacks.py:310] 2024-07-16 17:18:18,292 >> {'loss': 0.0013, 'learning_rate': 6.1323e-07, 'epoch': 3.85, 'throughput': 620.61}
+
+[INFO|callbacks.py:310] 2024-07-16 17:18:29,277 >> {'loss': 0.0003, 'learning_rate': 5.8489e-07, 'epoch': 3.87, 'throughput': 620.76}
+
+[INFO|callbacks.py:310] 2024-07-16 17:18:40,277 >> {'loss': 0.0026, 'learning_rate': 5.5714e-07, 'epoch': 3.90, 'throughput': 620.61}
+
+[INFO|callbacks.py:310] 2024-07-16 17:18:51,272 >> {'loss': 0.0097, 'learning_rate': 5.2997e-07, 'epoch': 3.92, 'throughput': 620.67}
+
+[INFO|callbacks.py:310] 2024-07-16 17:19:02,251 >> {'loss': 0.0047, 'learning_rate': 5.0341e-07, 'epoch': 3.95, 'throughput': 620.62}
+
+[INFO|callbacks.py:310] 2024-07-16 17:19:13,213 >> {'loss': 0.0081, 'learning_rate': 4.7746e-07, 'epoch': 3.97, 'throughput': 620.72}
+
+[INFO|callbacks.py:310] 2024-07-16 17:19:24,189 >> {'loss': 0.0018, 'learning_rate': 4.5212e-07, 'epoch': 4.00, 'throughput': 620.95}
+
+[INFO|callbacks.py:310] 2024-07-16 17:19:35,181 >> {'loss': 0.0053, 'learning_rate': 4.2741e-07, 'epoch': 4.03, 'throughput': 620.97}
+
+[INFO|callbacks.py:310] 2024-07-16 17:19:46,170 >> {'loss': 0.0005, 'learning_rate': 4.0332e-07, 'epoch': 4.05, 'throughput': 621.01}
+
+[INFO|callbacks.py:310] 2024-07-16 17:19:57,184 >> {'loss': 0.0001, 'learning_rate': 3.7988e-07, 'epoch': 4.08, 'throughput': 620.88}
+
+[INFO|callbacks.py:310] 2024-07-16 17:20:08,184 >> {'loss': 0.0018, 'learning_rate': 3.5708e-07, 'epoch': 4.10, 'throughput': 620.79}
+
+[INFO|callbacks.py:310] 2024-07-16 17:20:19,180 >> {'loss': 0.0010, 'learning_rate': 3.3494e-07, 'epoch': 4.13, 'throughput': 620.68}
+
+[INFO|callbacks.py:310] 2024-07-16 17:20:30,173 >> {'loss': 0.0001, 'learning_rate': 3.1345e-07, 'epoch': 4.15, 'throughput': 620.79}
+
+[INFO|callbacks.py:310] 2024-07-16 17:20:41,136 >> {'loss': 0.0012, 'learning_rate': 2.9263e-07, 'epoch': 4.18, 'throughput': 620.89}
+
+[INFO|callbacks.py:310] 2024-07-16 17:20:52,115 >> {'loss': 0.0001, 'learning_rate': 2.7248e-07, 'epoch': 4.21, 'throughput': 620.82}
+
+[INFO|callbacks.py:310] 2024-07-16 17:21:03,096 >> {'loss': 0.0002, 'learning_rate': 2.5301e-07, 'epoch': 4.23, 'throughput': 620.76}
+
+[INFO|callbacks.py:310] 2024-07-16 17:21:14,072 >> {'loss': 0.0001, 'learning_rate': 2.3423e-07, 'epoch': 4.26, 'throughput': 620.84}
+
+[INFO|callbacks.py:310] 2024-07-16 17:21:25,062 >> {'loss': 0.0003, 'learning_rate': 2.1614e-07, 'epoch': 4.28, 'throughput': 620.65}
+
+[INFO|callbacks.py:310] 2024-07-16 17:21:36,062 >> {'loss': 0.0001, 'learning_rate': 1.9874e-07, 'epoch': 4.31, 'throughput': 620.79}
+
+[INFO|callbacks.py:310] 2024-07-16 17:21:47,070 >> {'loss': 0.0007, 'learning_rate': 1.8204e-07, 'epoch': 4.34, 'throughput': 620.70}
+
+[INFO|callbacks.py:310] 2024-07-16 17:21:58,075 >> {'loss': 0.0003, 'learning_rate': 1.6605e-07, 'epoch': 4.36, 'throughput': 620.56}
+
+[INFO|callbacks.py:310] 2024-07-16 17:22:09,064 >> {'loss': 0.0001, 'learning_rate': 1.5077e-07, 'epoch': 4.39, 'throughput': 620.70}
+
+[INFO|callbacks.py:310] 2024-07-16 17:22:20,044 >> {'loss': 0.0008, 'learning_rate': 1.3620e-07, 'epoch': 4.41, 'throughput': 620.90}
+
+[INFO|callbacks.py:310] 2024-07-16 17:22:31,005 >> {'loss': 0.0001, 'learning_rate': 1.2236e-07, 'epoch': 4.44, 'throughput': 620.92}
+
+[INFO|callbacks.py:310] 2024-07-16 17:22:41,991 >> {'loss': 0.0004, 'learning_rate': 1.0924e-07, 'epoch': 4.46, 'throughput': 621.02}
+
+[INFO|callbacks.py:310] 2024-07-16 17:22:52,983 >> {'loss': 0.0001, 'learning_rate': 9.6846e-08, 'epoch': 4.49, 'throughput': 620.99}
+
+[INFO|callbacks.py:310] 2024-07-16 17:23:03,992 >> {'loss': 0.0005, 'learning_rate': 8.5185e-08, 'epoch': 4.52, 'throughput': 621.03}
+
+[INFO|callbacks.py:310] 2024-07-16 17:23:14,988 >> {'loss': 0.0076, 'learning_rate': 7.4261e-08, 'epoch': 4.54, 'throughput': 621.25}
+
+[INFO|callbacks.py:310] 2024-07-16 17:23:26,004 >> {'loss': 0.0004, 'learning_rate': 6.4075e-08, 'epoch': 4.57, 'throughput': 621.12}
+
+[INFO|callbacks.py:310] 2024-07-16 17:23:37,007 >> {'loss': 0.0040, 'learning_rate': 5.4631e-08, 'epoch': 4.59, 'throughput': 621.16}
+
+[INFO|callbacks.py:310] 2024-07-16 17:23:47,992 >> {'loss': 0.0001, 'learning_rate': 4.5932e-08, 'epoch': 4.62, 'throughput': 621.22}
+
+[INFO|callbacks.py:310] 2024-07-16 17:23:58,970 >> {'loss': 0.0005, 'learning_rate': 3.7981e-08, 'epoch': 4.65, 'throughput': 621.17}
+
+[INFO|callbacks.py:310] 2024-07-16 17:24:09,940 >> {'loss': 0.0002, 'learning_rate': 3.0779e-08, 'epoch': 4.67, 'throughput': 621.23}
+
+[INFO|callbacks.py:310] 2024-07-16 17:24:20,909 >> {'loss': 0.0001, 'learning_rate': 2.4330e-08, 'epoch': 4.70, 'throughput': 621.25}
+
+[INFO|callbacks.py:310] 2024-07-16 17:24:31,898 >> {'loss': 0.0001, 'learning_rate': 1.8635e-08, 'epoch': 4.72, 'throughput': 621.28}
+
+[INFO|callbacks.py:310] 2024-07-16 17:24:42,890 >> {'loss': 0.0001, 'learning_rate': 1.3695e-08, 'epoch': 4.75, 'throughput': 621.30}
+
+[INFO|callbacks.py:310] 2024-07-16 17:24:53,884 >> {'loss': 0.0167, 'learning_rate': 9.5133e-09, 'epoch': 4.77, 'throughput': 621.38}
+
+[INFO|callbacks.py:310] 2024-07-16 17:25:04,898 >> {'loss': 0.0002, 'learning_rate': 6.0899e-09, 'epoch': 4.80, 'throughput': 621.28}
+
+[INFO|callbacks.py:310] 2024-07-16 17:25:15,899 >> {'loss': 0.0003, 'learning_rate': 3.4262e-09, 'epoch': 4.83, 'throughput': 621.31}
+
+[INFO|callbacks.py:310] 2024-07-16 17:25:26,879 >> {'loss': 0.0024, 'learning_rate': 1.5229e-09, 'epoch': 4.85, 'throughput': 621.29}
+
+[INFO|callbacks.py:310] 2024-07-16 17:25:37,870 >> {'loss': 0.0005, 'learning_rate': 3.8076e-10, 'epoch': 4.88, 'throughput': 621.19}
+
+[INFO|callbacks.py:310] 2024-07-16 17:25:48,849 >> {'loss': 0.0017, 'learning_rate': 0.0000e+00, 'epoch': 4.90, 'throughput': 621.15}
+
+[INFO|trainer.py:3478] 2024-07-16 17:25:55,242 >> Saving model checkpoint to saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/checkpoint-190
+
+[INFO|configuration_utils.py:472] 2024-07-16 17:25:55,245 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/checkpoint-190/config.json
+
+[INFO|configuration_utils.py:769] 2024-07-16 17:25:55,246 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/checkpoint-190/generation_config.json
+
+[INFO|modeling_utils.py:2698] 2024-07-16 17:26:08,802 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 3 checkpoint shards. You can find where each parameters has been saved in the index located at saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/checkpoint-190/model.safetensors.index.json.
+
+[INFO|tokenization_utils_base.py:2574] 2024-07-16 17:26:08,803 >> tokenizer config file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/checkpoint-190/tokenizer_config.json
+
+[INFO|tokenization_utils_base.py:2583] 2024-07-16 17:26:08,803 >> Special tokens file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/checkpoint-190/special_tokens_map.json
+
+[INFO|trainer.py:2383] 2024-07-16 17:26:40,079 >> 
+
+Training completed. Do not forget to share your model on huggingface.co/models =)
+
+
+
+[INFO|trainer.py:3478] 2024-07-16 17:26:46,683 >> Saving model checkpoint to saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2
+
+[INFO|configuration_utils.py:472] 2024-07-16 17:26:46,686 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/config.json
+
+[INFO|configuration_utils.py:769] 2024-07-16 17:26:46,686 >> Configuration saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/generation_config.json
+
+[INFO|modeling_utils.py:2698] 2024-07-16 17:27:01,023 >> The model is bigger than the maximum size per checkpoint (5GB) and is going to be split in 3 checkpoint shards. You can find where each parameters has been saved in the index located at saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/model.safetensors.index.json.
+
+[INFO|tokenization_utils_base.py:2574] 2024-07-16 17:27:01,024 >> tokenizer config file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/tokenizer_config.json
+
+[INFO|tokenization_utils_base.py:2583] 2024-07-16 17:27:01,025 >> Special tokens file saved in saves/LLaMA2-7B-Chat/full/train_2024-07-16-16-48-49_llama2_2/special_tokens_map.json
+
+[WARNING|ploting.py:89] 2024-07-16 17:27:02,103 >> No metric eval_loss to plot.
+
+[WARNING|ploting.py:89] 2024-07-16 17:27:02,103 >> No metric eval_accuracy to plot.
+
+[INFO|modelcard.py:449] 2024-07-16 17:27:02,103 >> Dropping the following result as it does not have all the necessary fields:
+{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
+

special_tokens_map.json

@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723

tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if loop.index0 == 0 and system_message is defined %}{% set content = '<<SYS>>\n' + system_message + '\n<</SYS>>\n\n' + message['content'] %}{% endif %}{% if message['role'] == 'user' %}{{ '<s>' + '[INST] ' + content + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ content + '</s>' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 4.903225806451613,
+    "num_input_tokens_seen": 1299120,
+    "total_flos": 5.150239379108659e+16,
+    "train_loss": 0.3450760219912857,
+    "train_runtime": 2142.6975,
+    "train_samples_per_second": 11.57,
+    "train_steps_per_second": 0.089
+}

trainer_log.jsonl

@@ -0,0 +1,191 @@
+{"current_steps": 1, "total_steps": 190, "loss": 8.2514, "learning_rate": 5.000000000000001e-07, "epoch": 0.025806451612903226, "percentage": 0.53, "elapsed_time": "0:00:12", "remaining_time": "0:39:32", "throughput": "545.43", "total_tokens": 6848}
+{"current_steps": 2, "total_steps": 190, "loss": 8.2793, "learning_rate": 1.0000000000000002e-06, "epoch": 0.05161290322580645, "percentage": 1.05, "elapsed_time": "0:00:23", "remaining_time": "0:36:55", "throughput": "584.51", "total_tokens": 13776}
+{"current_steps": 3, "total_steps": 190, "loss": 8.17, "learning_rate": 1.5e-06, "epoch": 0.07741935483870968, "percentage": 1.58, "elapsed_time": "0:00:34", "remaining_time": "0:35:54", "throughput": "598.15", "total_tokens": 20672}
+{"current_steps": 4, "total_steps": 190, "loss": 7.6197, "learning_rate": 2.0000000000000003e-06, "epoch": 0.1032258064516129, "percentage": 2.11, "elapsed_time": "0:00:45", "remaining_time": "0:35:17", "throughput": "609.21", "total_tokens": 27744}
+{"current_steps": 5, "total_steps": 190, "loss": 6.9491, "learning_rate": 2.5e-06, "epoch": 0.12903225806451613, "percentage": 2.63, "elapsed_time": "0:00:56", "remaining_time": "0:34:51", "throughput": "612.41", "total_tokens": 34624}
+{"current_steps": 6, "total_steps": 190, "loss": 5.2054, "learning_rate": 3e-06, "epoch": 0.15483870967741936, "percentage": 3.16, "elapsed_time": "0:01:07", "remaining_time": "0:34:31", "throughput": "613.36", "total_tokens": 41424}
+{"current_steps": 7, "total_steps": 190, "loss": 4.8642, "learning_rate": 3.5e-06, "epoch": 0.18064516129032257, "percentage": 3.68, "elapsed_time": "0:01:18", "remaining_time": "0:34:13", "throughput": "615.05", "total_tokens": 48304}
+{"current_steps": 8, "total_steps": 190, "loss": 3.2874, "learning_rate": 4.000000000000001e-06, "epoch": 0.2064516129032258, "percentage": 4.21, "elapsed_time": "0:01:29", "remaining_time": "0:33:57", "throughput": "615.94", "total_tokens": 55152}
+{"current_steps": 9, "total_steps": 190, "loss": 2.631, "learning_rate": 4.5e-06, "epoch": 0.23225806451612904, "percentage": 4.74, "elapsed_time": "0:01:40", "remaining_time": "0:33:42", "throughput": "613.25", "total_tokens": 61680}
+{"current_steps": 10, "total_steps": 190, "loss": 0.6982, "learning_rate": 5e-06, "epoch": 0.25806451612903225, "percentage": 5.26, "elapsed_time": "0:01:51", "remaining_time": "0:33:28", "throughput": "613.59", "total_tokens": 68480}
+{"current_steps": 11, "total_steps": 190, "loss": 0.3276, "learning_rate": 4.9996192378909785e-06, "epoch": 0.2838709677419355, "percentage": 5.79, "elapsed_time": "0:02:02", "remaining_time": "0:33:15", "throughput": "613.98", "total_tokens": 75296}
+{"current_steps": 12, "total_steps": 190, "loss": 0.293, "learning_rate": 4.99847706754774e-06, "epoch": 0.3096774193548387, "percentage": 6.32, "elapsed_time": "0:02:13", "remaining_time": "0:33:02", "throughput": "615.72", "total_tokens": 82288}
+{"current_steps": 13, "total_steps": 190, "loss": 0.2129, "learning_rate": 4.9965738368864345e-06, "epoch": 0.33548387096774196, "percentage": 6.84, "elapsed_time": "0:02:24", "remaining_time": "0:32:49", "throughput": "615.94", "total_tokens": 89088}
+{"current_steps": 14, "total_steps": 190, "loss": 0.4712, "learning_rate": 4.993910125649561e-06, "epoch": 0.36129032258064514, "percentage": 7.37, "elapsed_time": "0:02:35", "remaining_time": "0:32:36", "throughput": "616.20", "total_tokens": 95904}
+{"current_steps": 15, "total_steps": 190, "loss": 0.235, "learning_rate": 4.990486745229364e-06, "epoch": 0.3870967741935484, "percentage": 7.89, "elapsed_time": "0:02:46", "remaining_time": "0:32:23", "throughput": "617.55", "total_tokens": 102896}
+{"current_steps": 16, "total_steps": 190, "loss": 0.202, "learning_rate": 4.986304738420684e-06, "epoch": 0.4129032258064516, "percentage": 8.42, "elapsed_time": "0:02:57", "remaining_time": "0:32:11", "throughput": "618.05", "total_tokens": 109792}
+{"current_steps": 17, "total_steps": 190, "loss": 0.1981, "learning_rate": 4.981365379103306e-06, "epoch": 0.43870967741935485, "percentage": 8.95, "elapsed_time": "0:03:08", "remaining_time": "0:31:59", "throughput": "617.73", "total_tokens": 116528}
+{"current_steps": 18, "total_steps": 190, "loss": 0.1517, "learning_rate": 4.975670171853926e-06, "epoch": 0.4645161290322581, "percentage": 9.47, "elapsed_time": "0:03:19", "remaining_time": "0:31:47", "throughput": "617.10", "total_tokens": 123200}
+{"current_steps": 19, "total_steps": 190, "loss": 0.4335, "learning_rate": 4.9692208514878445e-06, "epoch": 0.49032258064516127, "percentage": 10.0, "elapsed_time": "0:03:30", "remaining_time": "0:31:35", "throughput": "617.28", "total_tokens": 130032}
+{"current_steps": 20, "total_steps": 190, "loss": 0.3609, "learning_rate": 4.962019382530521e-06, "epoch": 0.5161290322580645, "percentage": 10.53, "elapsed_time": "0:03:41", "remaining_time": "0:31:24", "throughput": "617.29", "total_tokens": 136832}
+{"current_steps": 21, "total_steps": 190, "loss": 0.1708, "learning_rate": 4.9540679586191605e-06, "epoch": 0.5419354838709678, "percentage": 11.05, "elapsed_time": "0:03:52", "remaining_time": "0:31:12", "throughput": "618.54", "total_tokens": 143904}
+{"current_steps": 22, "total_steps": 190, "loss": 0.2277, "learning_rate": 4.9453690018345144e-06, "epoch": 0.567741935483871, "percentage": 11.58, "elapsed_time": "0:04:03", "remaining_time": "0:31:00", "throughput": "617.70", "total_tokens": 150496}
+{"current_steps": 23, "total_steps": 190, "loss": 0.3437, "learning_rate": 4.935925161963089e-06, "epoch": 0.5935483870967742, "percentage": 12.11, "elapsed_time": "0:04:14", "remaining_time": "0:30:49", "throughput": "617.93", "total_tokens": 157360}
+{"current_steps": 24, "total_steps": 190, "loss": 0.2229, "learning_rate": 4.925739315689991e-06, "epoch": 0.6193548387096774, "percentage": 12.63, "elapsed_time": "0:04:25", "remaining_time": "0:30:37", "throughput": "619.02", "total_tokens": 164464}
+{"current_steps": 25, "total_steps": 190, "loss": 0.1242, "learning_rate": 4.914814565722671e-06, "epoch": 0.6451612903225806, "percentage": 13.16, "elapsed_time": "0:04:36", "remaining_time": "0:30:26", "throughput": "617.82", "total_tokens": 170960}
+{"current_steps": 26, "total_steps": 190, "loss": 0.2117, "learning_rate": 4.903154239845798e-06, "epoch": 0.6709677419354839, "percentage": 13.68, "elapsed_time": "0:04:47", "remaining_time": "0:30:15", "throughput": "617.94", "total_tokens": 177808}
+{"current_steps": 27, "total_steps": 190, "loss": 0.2706, "learning_rate": 4.890761889907589e-06, "epoch": 0.6967741935483871, "percentage": 14.21, "elapsed_time": "0:04:58", "remaining_time": "0:30:03", "throughput": "618.70", "total_tokens": 184848}
+{"current_steps": 28, "total_steps": 190, "loss": 0.2084, "learning_rate": 4.8776412907378845e-06, "epoch": 0.7225806451612903, "percentage": 14.74, "elapsed_time": "0:05:09", "remaining_time": "0:29:52", "throughput": "618.27", "total_tokens": 191536}
+{"current_steps": 29, "total_steps": 190, "loss": 0.0981, "learning_rate": 4.863796438998293e-06, "epoch": 0.7483870967741936, "percentage": 15.26, "elapsed_time": "0:05:20", "remaining_time": "0:29:40", "throughput": "618.39", "total_tokens": 198368}
+{"current_steps": 30, "total_steps": 190, "loss": 0.16, "learning_rate": 4.849231551964771e-06, "epoch": 0.7741935483870968, "percentage": 15.79, "elapsed_time": "0:05:31", "remaining_time": "0:29:29", "throughput": "618.50", "total_tokens": 205200}
+{"current_steps": 31, "total_steps": 190, "loss": 0.1614, "learning_rate": 4.833951066243004e-06, "epoch": 0.8, "percentage": 16.32, "elapsed_time": "0:05:42", "remaining_time": "0:29:18", "throughput": "617.80", "total_tokens": 211760}
+{"current_steps": 32, "total_steps": 190, "loss": 0.1742, "learning_rate": 4.817959636416969e-06, "epoch": 0.8258064516129032, "percentage": 16.84, "elapsed_time": "0:05:53", "remaining_time": "0:29:06", "throughput": "617.47", "total_tokens": 218448}
+{"current_steps": 33, "total_steps": 190, "loss": 0.1107, "learning_rate": 4.801262133631101e-06, "epoch": 0.8516129032258064, "percentage": 17.37, "elapsed_time": "0:06:04", "remaining_time": "0:28:55", "throughput": "617.90", "total_tokens": 225408}
+{"current_steps": 34, "total_steps": 190, "loss": 0.0822, "learning_rate": 4.783863644106502e-06, "epoch": 0.8774193548387097, "percentage": 17.89, "elapsed_time": "0:06:15", "remaining_time": "0:28:44", "throughput": "617.42", "total_tokens": 232032}
+{"current_steps": 35, "total_steps": 190, "loss": 0.1873, "learning_rate": 4.765769467591626e-06, "epoch": 0.9032258064516129, "percentage": 18.42, "elapsed_time": "0:06:26", "remaining_time": "0:28:33", "throughput": "617.01", "total_tokens": 238672}
+{"current_steps": 36, "total_steps": 190, "loss": 0.2375, "learning_rate": 4.746985115747918e-06, "epoch": 0.9290322580645162, "percentage": 18.95, "elapsed_time": "0:06:37", "remaining_time": "0:28:21", "throughput": "616.94", "total_tokens": 245456}
+{"current_steps": 37, "total_steps": 190, "loss": 0.2667, "learning_rate": 4.72751631047092e-06, "epoch": 0.9548387096774194, "percentage": 19.47, "elapsed_time": "0:06:48", "remaining_time": "0:28:10", "throughput": "617.73", "total_tokens": 252576}
+{"current_steps": 38, "total_steps": 190, "loss": 0.1547, "learning_rate": 4.707368982147318e-06, "epoch": 0.9806451612903225, "percentage": 20.0, "elapsed_time": "0:06:59", "remaining_time": "0:27:59", "throughput": "618.14", "total_tokens": 259536}
+{"current_steps": 39, "total_steps": 190, "loss": 0.1662, "learning_rate": 4.68654926784849e-06, "epoch": 1.0064516129032257, "percentage": 20.53, "elapsed_time": "0:07:10", "remaining_time": "0:27:48", "throughput": "618.69", "total_tokens": 266560}
+{"current_steps": 40, "total_steps": 190, "loss": 0.0808, "learning_rate": 4.665063509461098e-06, "epoch": 1.032258064516129, "percentage": 21.05, "elapsed_time": "0:07:21", "remaining_time": "0:27:36", "throughput": "618.41", "total_tokens": 273248}
+{"current_steps": 41, "total_steps": 190, "loss": 0.0884, "learning_rate": 4.642918251755281e-06, "epoch": 1.0580645161290323, "percentage": 21.58, "elapsed_time": "0:07:32", "remaining_time": "0:27:25", "throughput": "618.04", "total_tokens": 279888}
+{"current_steps": 42, "total_steps": 190, "loss": 0.0883, "learning_rate": 4.620120240391065e-06, "epoch": 1.0838709677419356, "percentage": 22.11, "elapsed_time": "0:07:43", "remaining_time": "0:27:14", "throughput": "618.55", "total_tokens": 286944}
+{"current_steps": 43, "total_steps": 190, "loss": 0.0562, "learning_rate": 4.596676419863561e-06, "epoch": 1.1096774193548387, "percentage": 22.63, "elapsed_time": "0:07:54", "remaining_time": "0:27:03", "throughput": "618.59", "total_tokens": 293792}
+{"current_steps": 44, "total_steps": 190, "loss": 0.0856, "learning_rate": 4.572593931387604e-06, "epoch": 1.135483870967742, "percentage": 23.16, "elapsed_time": "0:08:05", "remaining_time": "0:26:52", "throughput": "618.38", "total_tokens": 300512}
+{"current_steps": 45, "total_steps": 190, "loss": 0.0612, "learning_rate": 4.54788011072248e-06, "epoch": 1.1612903225806452, "percentage": 23.68, "elapsed_time": "0:08:16", "remaining_time": "0:26:41", "throughput": "618.26", "total_tokens": 307264}
+{"current_steps": 46, "total_steps": 190, "loss": 0.0944, "learning_rate": 4.522542485937369e-06, "epoch": 1.1870967741935483, "percentage": 24.21, "elapsed_time": "0:08:27", "remaining_time": "0:26:30", "throughput": "618.35", "total_tokens": 314112}
+{"current_steps": 47, "total_steps": 190, "loss": 0.0624, "learning_rate": 4.496588775118232e-06, "epoch": 1.2129032258064516, "percentage": 24.74, "elapsed_time": "0:08:39", "remaining_time": "0:26:19", "throughput": "618.23", "total_tokens": 320864}
+{"current_steps": 48, "total_steps": 190, "loss": 0.0363, "learning_rate": 4.470026884016805e-06, "epoch": 1.238709677419355, "percentage": 25.26, "elapsed_time": "0:08:49", "remaining_time": "0:26:07", "throughput": "618.21", "total_tokens": 327648}
+{"current_steps": 49, "total_steps": 190, "loss": 0.1039, "learning_rate": 4.442864903642428e-06, "epoch": 1.2645161290322582, "percentage": 25.79, "elapsed_time": "0:09:01", "remaining_time": "0:25:56", "throughput": "618.16", "total_tokens": 334432}
+{"current_steps": 50, "total_steps": 190, "loss": 0.0488, "learning_rate": 4.415111107797445e-06, "epoch": 1.2903225806451613, "percentage": 26.32, "elapsed_time": "0:09:12", "remaining_time": "0:25:45", "throughput": "618.19", "total_tokens": 341248}
+{"current_steps": 51, "total_steps": 190, "loss": 0.0613, "learning_rate": 4.386773950556931e-06, "epoch": 1.3161290322580645, "percentage": 26.84, "elapsed_time": "0:09:23", "remaining_time": "0:25:34", "throughput": "618.44", "total_tokens": 348192}
+{"current_steps": 52, "total_steps": 190, "loss": 0.07, "learning_rate": 4.357862063693486e-06, "epoch": 1.3419354838709676, "percentage": 27.37, "elapsed_time": "0:09:34", "remaining_time": "0:25:23", "throughput": "618.12", "total_tokens": 354816}
+{"current_steps": 53, "total_steps": 190, "loss": 0.0463, "learning_rate": 4.328384254047927e-06, "epoch": 1.367741935483871, "percentage": 27.89, "elapsed_time": "0:09:45", "remaining_time": "0:25:12", "throughput": "618.08", "total_tokens": 361600}
+{"current_steps": 54, "total_steps": 190, "loss": 0.0671, "learning_rate": 4.2983495008466285e-06, "epoch": 1.3935483870967742, "percentage": 28.42, "elapsed_time": "0:09:56", "remaining_time": "0:25:01", "throughput": "618.15", "total_tokens": 368464}
+{"current_steps": 55, "total_steps": 190, "loss": 0.0428, "learning_rate": 4.267766952966369e-06, "epoch": 1.4193548387096775, "percentage": 28.95, "elapsed_time": "0:10:07", "remaining_time": "0:24:50", "throughput": "618.43", "total_tokens": 375440}
+{"current_steps": 56, "total_steps": 190, "loss": 0.0678, "learning_rate": 4.236645926147493e-06, "epoch": 1.4451612903225808, "percentage": 29.47, "elapsed_time": "0:10:18", "remaining_time": "0:24:38", "throughput": "618.43", "total_tokens": 382240}
+{"current_steps": 57, "total_steps": 190, "loss": 0.0476, "learning_rate": 4.204995900156247e-06, "epoch": 1.4709677419354839, "percentage": 30.0, "elapsed_time": "0:10:29", "remaining_time": "0:24:27", "throughput": "618.38", "total_tokens": 389008}
+{"current_steps": 58, "total_steps": 190, "loss": 0.0442, "learning_rate": 4.172826515897146e-06, "epoch": 1.4967741935483871, "percentage": 30.53, "elapsed_time": "0:10:40", "remaining_time": "0:24:16", "throughput": "618.82", "total_tokens": 396080}
+{"current_steps": 59, "total_steps": 190, "loss": 0.0336, "learning_rate": 4.140147572476269e-06, "epoch": 1.5225806451612902, "percentage": 31.05, "elapsed_time": "0:10:51", "remaining_time": "0:24:05", "throughput": "619.09", "total_tokens": 403056}
+{"current_steps": 60, "total_steps": 190, "loss": 0.046, "learning_rate": 4.106969024216348e-06, "epoch": 1.5483870967741935, "percentage": 31.58, "elapsed_time": "0:11:02", "remaining_time": "0:23:54", "throughput": "618.77", "total_tokens": 409664}
+{"current_steps": 61, "total_steps": 190, "loss": 0.0416, "learning_rate": 4.073300977624594e-06, "epoch": 1.5741935483870968, "percentage": 32.11, "elapsed_time": "0:11:13", "remaining_time": "0:23:43", "throughput": "618.46", "total_tokens": 416272}
+{"current_steps": 62, "total_steps": 190, "loss": 0.0649, "learning_rate": 4.039153688314146e-06, "epoch": 1.6, "percentage": 32.63, "elapsed_time": "0:11:24", "remaining_time": "0:23:32", "throughput": "618.87", "total_tokens": 423360}
+{"current_steps": 63, "total_steps": 190, "loss": 0.0591, "learning_rate": 4.0045375578801216e-06, "epoch": 1.6258064516129034, "percentage": 33.16, "elapsed_time": "0:11:35", "remaining_time": "0:23:21", "throughput": "619.05", "total_tokens": 430304}
+{"current_steps": 64, "total_steps": 190, "loss": 0.0318, "learning_rate": 3.969463130731183e-06, "epoch": 1.6516129032258065, "percentage": 33.68, "elapsed_time": "0:11:46", "remaining_time": "0:23:10", "throughput": "618.83", "total_tokens": 436960}
+{"current_steps": 65, "total_steps": 190, "loss": 0.0462, "learning_rate": 3.933941090877615e-06, "epoch": 1.6774193548387095, "percentage": 34.21, "elapsed_time": "0:11:57", "remaining_time": "0:22:59", "throughput": "618.87", "total_tokens": 443792}
+{"current_steps": 66, "total_steps": 190, "loss": 0.0465, "learning_rate": 3.897982258676867e-06, "epoch": 1.7032258064516128, "percentage": 34.74, "elapsed_time": "0:12:08", "remaining_time": "0:22:47", "throughput": "618.98", "total_tokens": 450672}
+{"current_steps": 67, "total_steps": 190, "loss": 0.0316, "learning_rate": 3.861597587537568e-06, "epoch": 1.729032258064516, "percentage": 35.26, "elapsed_time": "0:12:19", "remaining_time": "0:22:36", "throughput": "619.15", "total_tokens": 457616}
+{"current_steps": 68, "total_steps": 190, "loss": 0.1, "learning_rate": 3.824798160583012e-06, "epoch": 1.7548387096774194, "percentage": 35.79, "elapsed_time": "0:12:30", "remaining_time": "0:22:25", "throughput": "619.38", "total_tokens": 464592}
+{"current_steps": 69, "total_steps": 190, "loss": 0.0711, "learning_rate": 3.787595187275136e-06, "epoch": 1.7806451612903227, "percentage": 36.32, "elapsed_time": "0:12:41", "remaining_time": "0:22:14", "throughput": "619.25", "total_tokens": 471328}
+{"current_steps": 70, "total_steps": 190, "loss": 0.0494, "learning_rate": 3.7500000000000005e-06, "epoch": 1.8064516129032258, "percentage": 36.84, "elapsed_time": "0:12:52", "remaining_time": "0:22:03", "throughput": "619.69", "total_tokens": 478480}
+{"current_steps": 71, "total_steps": 190, "loss": 0.0618, "learning_rate": 3.7120240506158433e-06, "epoch": 1.832258064516129, "percentage": 37.37, "elapsed_time": "0:13:03", "remaining_time": "0:21:52", "throughput": "619.64", "total_tokens": 485264}
+{"current_steps": 72, "total_steps": 190, "loss": 0.0511, "learning_rate": 3.6736789069647273e-06, "epoch": 1.8580645161290321, "percentage": 37.89, "elapsed_time": "0:13:14", "remaining_time": "0:21:41", "throughput": "619.61", "total_tokens": 492064}
+{"current_steps": 73, "total_steps": 190, "loss": 0.0464, "learning_rate": 3.634976249348867e-06, "epoch": 1.8838709677419354, "percentage": 38.42, "elapsed_time": "0:13:25", "remaining_time": "0:21:30", "throughput": "619.64", "total_tokens": 498896}
+{"current_steps": 74, "total_steps": 190, "loss": 0.0331, "learning_rate": 3.595927866972694e-06, "epoch": 1.9096774193548387, "percentage": 38.95, "elapsed_time": "0:13:36", "remaining_time": "0:21:19", "throughput": "620.02", "total_tokens": 506016}
+{"current_steps": 75, "total_steps": 190, "loss": 0.0706, "learning_rate": 3.556545654351749e-06, "epoch": 1.935483870967742, "percentage": 39.47, "elapsed_time": "0:13:47", "remaining_time": "0:21:08", "throughput": "620.04", "total_tokens": 512848}
+{"current_steps": 76, "total_steps": 190, "loss": 0.0442, "learning_rate": 3.516841607689501e-06, "epoch": 1.9612903225806453, "percentage": 40.0, "elapsed_time": "0:13:58", "remaining_time": "0:20:57", "throughput": "620.14", "total_tokens": 519744}
+{"current_steps": 77, "total_steps": 190, "loss": 0.042, "learning_rate": 3.476827821223184e-06, "epoch": 1.9870967741935484, "percentage": 40.53, "elapsed_time": "0:14:09", "remaining_time": "0:20:46", "throughput": "619.89", "total_tokens": 526352}
+{"current_steps": 78, "total_steps": 190, "loss": 0.021, "learning_rate": 3.436516483539781e-06, "epoch": 2.0129032258064514, "percentage": 41.05, "elapsed_time": "0:14:20", "remaining_time": "0:20:35", "throughput": "619.84", "total_tokens": 533136}
+{"current_steps": 79, "total_steps": 190, "loss": 0.0094, "learning_rate": 3.39591987386325e-06, "epoch": 2.0387096774193547, "percentage": 41.58, "elapsed_time": "0:14:31", "remaining_time": "0:20:23", "throughput": "619.91", "total_tokens": 540016}
+{"current_steps": 80, "total_steps": 190, "loss": 0.0021, "learning_rate": 3.3550503583141726e-06, "epoch": 2.064516129032258, "percentage": 42.11, "elapsed_time": "0:14:42", "remaining_time": "0:20:12", "throughput": "620.26", "total_tokens": 547152}
+{"current_steps": 81, "total_steps": 190, "loss": 0.0146, "learning_rate": 3.313920386142892e-06, "epoch": 2.0903225806451613, "percentage": 42.63, "elapsed_time": "0:14:53", "remaining_time": "0:20:01", "throughput": "620.13", "total_tokens": 553888}
+{"current_steps": 82, "total_steps": 190, "loss": 0.0237, "learning_rate": 3.272542485937369e-06, "epoch": 2.1161290322580646, "percentage": 43.16, "elapsed_time": "0:15:04", "remaining_time": "0:19:50", "throughput": "620.48", "total_tokens": 561024}
+{"current_steps": 83, "total_steps": 190, "loss": 0.0031, "learning_rate": 3.230929261806842e-06, "epoch": 2.141935483870968, "percentage": 43.68, "elapsed_time": "0:15:15", "remaining_time": "0:19:39", "throughput": "620.57", "total_tokens": 567936}
+{"current_steps": 84, "total_steps": 190, "loss": 0.0034, "learning_rate": 3.189093389542498e-06, "epoch": 2.167741935483871, "percentage": 44.21, "elapsed_time": "0:15:26", "remaining_time": "0:19:28", "throughput": "620.54", "total_tokens": 574736}
+{"current_steps": 85, "total_steps": 190, "loss": 0.0045, "learning_rate": 3.147047612756302e-06, "epoch": 2.193548387096774, "percentage": 44.74, "elapsed_time": "0:15:37", "remaining_time": "0:19:17", "throughput": "620.66", "total_tokens": 581680}
+{"current_steps": 86, "total_steps": 190, "loss": 0.0031, "learning_rate": 3.1048047389991693e-06, "epoch": 2.2193548387096773, "percentage": 45.26, "elapsed_time": "0:15:48", "remaining_time": "0:19:06", "throughput": "620.62", "total_tokens": 588464}
+{"current_steps": 87, "total_steps": 190, "loss": 0.0341, "learning_rate": 3.062377635859663e-06, "epoch": 2.2451612903225806, "percentage": 45.79, "elapsed_time": "0:15:59", "remaining_time": "0:18:55", "throughput": "620.63", "total_tokens": 595312}
+{"current_steps": 88, "total_steps": 190, "loss": 0.0095, "learning_rate": 3.019779227044398e-06, "epoch": 2.270967741935484, "percentage": 46.32, "elapsed_time": "0:16:10", "remaining_time": "0:18:44", "throughput": "620.55", "total_tokens": 602080}
+{"current_steps": 89, "total_steps": 190, "loss": 0.0459, "learning_rate": 2.9770224884413625e-06, "epoch": 2.296774193548387, "percentage": 46.84, "elapsed_time": "0:16:21", "remaining_time": "0:18:33", "throughput": "620.80", "total_tokens": 609168}
+{"current_steps": 90, "total_steps": 190, "loss": 0.0104, "learning_rate": 2.9341204441673267e-06, "epoch": 2.3225806451612905, "percentage": 47.37, "elapsed_time": "0:16:32", "remaining_time": "0:18:22", "throughput": "620.80", "total_tokens": 616000}
+{"current_steps": 91, "total_steps": 190, "loss": 0.0201, "learning_rate": 2.8910861626005774e-06, "epoch": 2.3483870967741938, "percentage": 47.89, "elapsed_time": "0:16:43", "remaining_time": "0:18:11", "throughput": "620.51", "total_tokens": 622544}
+{"current_steps": 92, "total_steps": 190, "loss": 0.0021, "learning_rate": 2.847932752400164e-06, "epoch": 2.3741935483870966, "percentage": 48.42, "elapsed_time": "0:16:54", "remaining_time": "0:18:00", "throughput": "620.75", "total_tokens": 629600}
+{"current_steps": 93, "total_steps": 190, "loss": 0.043, "learning_rate": 2.804673358512869e-06, "epoch": 2.4, "percentage": 48.95, "elapsed_time": "0:17:05", "remaining_time": "0:17:49", "throughput": "620.64", "total_tokens": 636320}
+{"current_steps": 94, "total_steps": 190, "loss": 0.0207, "learning_rate": 2.761321158169134e-06, "epoch": 2.425806451612903, "percentage": 49.47, "elapsed_time": "0:17:16", "remaining_time": "0:17:38", "throughput": "620.61", "total_tokens": 643136}
+{"current_steps": 95, "total_steps": 190, "loss": 0.0148, "learning_rate": 2.717889356869146e-06, "epoch": 2.4516129032258065, "percentage": 50.0, "elapsed_time": "0:17:27", "remaining_time": "0:17:27", "throughput": "620.69", "total_tokens": 650048}
+{"current_steps": 96, "total_steps": 190, "loss": 0.004, "learning_rate": 2.6743911843603134e-06, "epoch": 2.47741935483871, "percentage": 50.53, "elapsed_time": "0:17:38", "remaining_time": "0:17:16", "throughput": "620.55", "total_tokens": 656736}
+{"current_steps": 97, "total_steps": 190, "loss": 0.0131, "learning_rate": 2.6308398906073603e-06, "epoch": 2.5032258064516126, "percentage": 51.05, "elapsed_time": "0:17:49", "remaining_time": "0:17:05", "throughput": "620.54", "total_tokens": 663552}
+{"current_steps": 98, "total_steps": 190, "loss": 0.0455, "learning_rate": 2.587248741756253e-06, "epoch": 2.5290322580645164, "percentage": 51.58, "elapsed_time": "0:18:00", "remaining_time": "0:16:54", "throughput": "620.37", "total_tokens": 670224}
+{"current_steps": 99, "total_steps": 190, "loss": 0.0031, "learning_rate": 2.543631016093209e-06, "epoch": 2.554838709677419, "percentage": 52.11, "elapsed_time": "0:18:11", "remaining_time": "0:16:43", "throughput": "620.27", "total_tokens": 676960}
+{"current_steps": 100, "total_steps": 190, "loss": 0.0099, "learning_rate": 2.5e-06, "epoch": 2.5806451612903225, "percentage": 52.63, "elapsed_time": "0:18:22", "remaining_time": "0:16:32", "throughput": "620.49", "total_tokens": 684016}
+{"current_steps": 101, "total_steps": 190, "loss": 0.0797, "learning_rate": 2.4563689839067913e-06, "epoch": 2.606451612903226, "percentage": 53.16, "elapsed_time": "0:18:33", "remaining_time": "0:16:21", "throughput": "620.49", "total_tokens": 690832}
+{"current_steps": 102, "total_steps": 190, "loss": 0.0059, "learning_rate": 2.4127512582437486e-06, "epoch": 2.632258064516129, "percentage": 53.68, "elapsed_time": "0:18:44", "remaining_time": "0:16:10", "throughput": "620.63", "total_tokens": 697808}
+{"current_steps": 103, "total_steps": 190, "loss": 0.0438, "learning_rate": 2.3691601093926406e-06, "epoch": 2.6580645161290324, "percentage": 54.21, "elapsed_time": "0:18:55", "remaining_time": "0:15:58", "throughput": "620.38", "total_tokens": 704352}
+{"current_steps": 104, "total_steps": 190, "loss": 0.0149, "learning_rate": 2.325608815639687e-06, "epoch": 2.6838709677419352, "percentage": 54.74, "elapsed_time": "0:19:06", "remaining_time": "0:15:47", "throughput": "620.63", "total_tokens": 711456}
+{"current_steps": 105, "total_steps": 190, "loss": 0.0126, "learning_rate": 2.2821106431308546e-06, "epoch": 2.709677419354839, "percentage": 55.26, "elapsed_time": "0:19:17", "remaining_time": "0:15:36", "throughput": "620.57", "total_tokens": 718224}
+{"current_steps": 106, "total_steps": 190, "loss": 0.0255, "learning_rate": 2.238678841830867e-06, "epoch": 2.735483870967742, "percentage": 55.79, "elapsed_time": "0:19:28", "remaining_time": "0:15:25", "throughput": "620.46", "total_tokens": 724928}
+{"current_steps": 107, "total_steps": 190, "loss": 0.0048, "learning_rate": 2.195326641487132e-06, "epoch": 2.761290322580645, "percentage": 56.32, "elapsed_time": "0:19:39", "remaining_time": "0:15:14", "throughput": "620.34", "total_tokens": 731616}
+{"current_steps": 108, "total_steps": 190, "loss": 0.0142, "learning_rate": 2.1520672475998374e-06, "epoch": 2.7870967741935484, "percentage": 56.84, "elapsed_time": "0:19:50", "remaining_time": "0:15:03", "throughput": "620.20", "total_tokens": 738272}
+{"current_steps": 109, "total_steps": 190, "loss": 0.0193, "learning_rate": 2.1089138373994226e-06, "epoch": 2.8129032258064517, "percentage": 57.37, "elapsed_time": "0:20:01", "remaining_time": "0:14:52", "throughput": "620.23", "total_tokens": 745104}
+{"current_steps": 110, "total_steps": 190, "loss": 0.0055, "learning_rate": 2.0658795558326745e-06, "epoch": 2.838709677419355, "percentage": 57.89, "elapsed_time": "0:20:12", "remaining_time": "0:14:41", "throughput": "620.32", "total_tokens": 752016}
+{"current_steps": 111, "total_steps": 190, "loss": 0.0144, "learning_rate": 2.022977511558638e-06, "epoch": 2.864516129032258, "percentage": 58.42, "elapsed_time": "0:20:23", "remaining_time": "0:14:30", "throughput": "620.23", "total_tokens": 758720}
+{"current_steps": 112, "total_steps": 190, "loss": 0.0272, "learning_rate": 1.9802207729556023e-06, "epoch": 2.8903225806451616, "percentage": 58.95, "elapsed_time": "0:20:34", "remaining_time": "0:14:19", "throughput": "620.22", "total_tokens": 765520}
+{"current_steps": 113, "total_steps": 190, "loss": 0.0101, "learning_rate": 1.937622364140338e-06, "epoch": 2.9161290322580644, "percentage": 59.47, "elapsed_time": "0:20:45", "remaining_time": "0:14:08", "throughput": "620.18", "total_tokens": 772288}
+{"current_steps": 114, "total_steps": 190, "loss": 0.0109, "learning_rate": 1.895195261000831e-06, "epoch": 2.9419354838709677, "percentage": 60.0, "elapsed_time": "0:20:56", "remaining_time": "0:13:57", "throughput": "620.51", "total_tokens": 779520}
+{"current_steps": 115, "total_steps": 190, "loss": 0.018, "learning_rate": 1.852952387243698e-06, "epoch": 2.967741935483871, "percentage": 60.53, "elapsed_time": "0:21:07", "remaining_time": "0:13:46", "throughput": "620.55", "total_tokens": 786400}
+{"current_steps": 116, "total_steps": 190, "loss": 0.0141, "learning_rate": 1.8109066104575023e-06, "epoch": 2.9935483870967743, "percentage": 61.05, "elapsed_time": "0:21:18", "remaining_time": "0:13:35", "throughput": "620.38", "total_tokens": 793008}
+{"current_steps": 117, "total_steps": 190, "loss": 0.0057, "learning_rate": 1.7690707381931585e-06, "epoch": 3.0193548387096776, "percentage": 61.58, "elapsed_time": "0:21:29", "remaining_time": "0:13:24", "throughput": "620.36", "total_tokens": 799808}
+{"current_steps": 118, "total_steps": 190, "loss": 0.0063, "learning_rate": 1.7274575140626318e-06, "epoch": 3.0451612903225804, "percentage": 62.11, "elapsed_time": "0:21:40", "remaining_time": "0:13:13", "throughput": "620.33", "total_tokens": 806576}
+{"current_steps": 119, "total_steps": 190, "loss": 0.0138, "learning_rate": 1.686079613857109e-06, "epoch": 3.0709677419354837, "percentage": 62.63, "elapsed_time": "0:21:51", "remaining_time": "0:13:02", "throughput": "620.27", "total_tokens": 813312}
+{"current_steps": 120, "total_steps": 190, "loss": 0.0011, "learning_rate": 1.6449496416858285e-06, "epoch": 3.096774193548387, "percentage": 63.16, "elapsed_time": "0:22:02", "remaining_time": "0:12:51", "throughput": "619.97", "total_tokens": 819728}
+{"current_steps": 121, "total_steps": 190, "loss": 0.0006, "learning_rate": 1.6040801261367494e-06, "epoch": 3.1225806451612903, "percentage": 63.68, "elapsed_time": "0:22:13", "remaining_time": "0:12:40", "throughput": "619.94", "total_tokens": 826496}
+{"current_steps": 122, "total_steps": 190, "loss": 0.0055, "learning_rate": 1.56348351646022e-06, "epoch": 3.1483870967741936, "percentage": 64.21, "elapsed_time": "0:22:24", "remaining_time": "0:12:29", "throughput": "619.97", "total_tokens": 833360}
+{"current_steps": 123, "total_steps": 190, "loss": 0.0011, "learning_rate": 1.5231721787768162e-06, "epoch": 3.174193548387097, "percentage": 64.74, "elapsed_time": "0:22:35", "remaining_time": "0:12:18", "throughput": "619.90", "total_tokens": 840080}
+{"current_steps": 124, "total_steps": 190, "loss": 0.0173, "learning_rate": 1.4831583923105e-06, "epoch": 3.2, "percentage": 65.26, "elapsed_time": "0:22:46", "remaining_time": "0:12:07", "throughput": "620.05", "total_tokens": 847104}
+{"current_steps": 125, "total_steps": 190, "loss": 0.0027, "learning_rate": 1.443454345648252e-06, "epoch": 3.225806451612903, "percentage": 65.79, "elapsed_time": "0:22:57", "remaining_time": "0:11:56", "throughput": "619.89", "total_tokens": 853712}
+{"current_steps": 126, "total_steps": 190, "loss": 0.0029, "learning_rate": 1.4040721330273063e-06, "epoch": 3.2516129032258063, "percentage": 66.32, "elapsed_time": "0:23:08", "remaining_time": "0:11:45", "throughput": "619.81", "total_tokens": 860400}
+{"current_steps": 127, "total_steps": 190, "loss": 0.0003, "learning_rate": 1.3650237506511333e-06, "epoch": 3.2774193548387096, "percentage": 66.84, "elapsed_time": "0:23:19", "remaining_time": "0:11:34", "throughput": "619.97", "total_tokens": 867440}
+{"current_steps": 128, "total_steps": 190, "loss": 0.0007, "learning_rate": 1.3263210930352737e-06, "epoch": 3.303225806451613, "percentage": 67.37, "elapsed_time": "0:23:30", "remaining_time": "0:11:23", "throughput": "619.98", "total_tokens": 874256}
+{"current_steps": 129, "total_steps": 190, "loss": 0.008, "learning_rate": 1.2879759493841577e-06, "epoch": 3.329032258064516, "percentage": 67.89, "elapsed_time": "0:23:41", "remaining_time": "0:11:11", "throughput": "620.05", "total_tokens": 881168}
+{"current_steps": 130, "total_steps": 190, "loss": 0.0004, "learning_rate": 1.2500000000000007e-06, "epoch": 3.3548387096774195, "percentage": 68.42, "elapsed_time": "0:23:52", "remaining_time": "0:11:00", "throughput": "620.16", "total_tokens": 888128}
+{"current_steps": 131, "total_steps": 190, "loss": 0.0049, "learning_rate": 1.2124048127248644e-06, "epoch": 3.3806451612903228, "percentage": 68.95, "elapsed_time": "0:24:03", "remaining_time": "0:10:49", "throughput": "620.39", "total_tokens": 895296}
+{"current_steps": 132, "total_steps": 190, "loss": 0.0012, "learning_rate": 1.1752018394169882e-06, "epoch": 3.4064516129032256, "percentage": 69.47, "elapsed_time": "0:24:14", "remaining_time": "0:10:38", "throughput": "620.30", "total_tokens": 901984}
+{"current_steps": 133, "total_steps": 190, "loss": 0.0044, "learning_rate": 1.1384024124624324e-06, "epoch": 3.432258064516129, "percentage": 70.0, "elapsed_time": "0:24:25", "remaining_time": "0:10:27", "throughput": "620.50", "total_tokens": 909104}
+{"current_steps": 134, "total_steps": 190, "loss": 0.0017, "learning_rate": 1.1020177413231334e-06, "epoch": 3.458064516129032, "percentage": 70.53, "elapsed_time": "0:24:36", "remaining_time": "0:10:16", "throughput": "620.57", "total_tokens": 916032}
+{"current_steps": 135, "total_steps": 190, "loss": 0.0003, "learning_rate": 1.0660589091223854e-06, "epoch": 3.4838709677419355, "percentage": 71.05, "elapsed_time": "0:24:47", "remaining_time": "0:10:05", "throughput": "620.49", "total_tokens": 922736}
+{"current_steps": 136, "total_steps": 190, "loss": 0.0099, "learning_rate": 1.0305368692688175e-06, "epoch": 3.509677419354839, "percentage": 71.58, "elapsed_time": "0:24:58", "remaining_time": "0:09:54", "throughput": "620.45", "total_tokens": 929472}
+{"current_steps": 137, "total_steps": 190, "loss": 0.0068, "learning_rate": 9.95462442119879e-07, "epoch": 3.535483870967742, "percentage": 72.11, "elapsed_time": "0:25:09", "remaining_time": "0:09:43", "throughput": "620.34", "total_tokens": 936112}
+{"current_steps": 138, "total_steps": 190, "loss": 0.0025, "learning_rate": 9.608463116858544e-07, "epoch": 3.5612903225806454, "percentage": 72.63, "elapsed_time": "0:25:20", "remaining_time": "0:09:32", "throughput": "620.33", "total_tokens": 942912}
+{"current_steps": 139, "total_steps": 190, "loss": 0.0004, "learning_rate": 9.266990223754069e-07, "epoch": 3.587096774193548, "percentage": 73.16, "elapsed_time": "0:25:30", "remaining_time": "0:09:21", "throughput": "620.50", "total_tokens": 949984}
+{"current_steps": 140, "total_steps": 190, "loss": 0.0101, "learning_rate": 8.930309757836517e-07, "epoch": 3.6129032258064515, "percentage": 73.68, "elapsed_time": "0:25:41", "remaining_time": "0:09:10", "throughput": "620.37", "total_tokens": 956608}
+{"current_steps": 141, "total_steps": 190, "loss": 0.0068, "learning_rate": 8.598524275237321e-07, "epoch": 3.638709677419355, "percentage": 74.21, "elapsed_time": "0:25:52", "remaining_time": "0:08:59", "throughput": "620.41", "total_tokens": 963488}
+{"current_steps": 142, "total_steps": 190, "loss": 0.0007, "learning_rate": 8.271734841028553e-07, "epoch": 3.664516129032258, "percentage": 74.74, "elapsed_time": "0:26:03", "remaining_time": "0:08:48", "throughput": "620.29", "total_tokens": 970128}
+{"current_steps": 143, "total_steps": 190, "loss": 0.0161, "learning_rate": 7.950040998437541e-07, "epoch": 3.6903225806451614, "percentage": 75.26, "elapsed_time": "0:26:14", "remaining_time": "0:08:37", "throughput": "620.19", "total_tokens": 976800}
+{"current_steps": 144, "total_steps": 190, "loss": 0.0115, "learning_rate": 7.633540738525066e-07, "epoch": 3.7161290322580647, "percentage": 75.79, "elapsed_time": "0:26:25", "remaining_time": "0:08:26", "throughput": "620.44", "total_tokens": 984000}
+{"current_steps": 145, "total_steps": 190, "loss": 0.0052, "learning_rate": 7.322330470336314e-07, "epoch": 3.741935483870968, "percentage": 76.32, "elapsed_time": "0:26:36", "remaining_time": "0:08:15", "throughput": "620.49", "total_tokens": 990896}
+{"current_steps": 146, "total_steps": 190, "loss": 0.0098, "learning_rate": 7.016504991533727e-07, "epoch": 3.767741935483871, "percentage": 76.84, "elapsed_time": "0:26:47", "remaining_time": "0:08:04", "throughput": "620.56", "total_tokens": 997824}
+{"current_steps": 147, "total_steps": 190, "loss": 0.0005, "learning_rate": 6.716157459520739e-07, "epoch": 3.793548387096774, "percentage": 77.37, "elapsed_time": "0:26:58", "remaining_time": "0:07:53", "throughput": "620.74", "total_tokens": 1004928}
+{"current_steps": 148, "total_steps": 190, "loss": 0.0012, "learning_rate": 6.421379363065142e-07, "epoch": 3.8193548387096774, "percentage": 77.89, "elapsed_time": "0:27:09", "remaining_time": "0:07:42", "throughput": "620.71", "total_tokens": 1011696}
+{"current_steps": 149, "total_steps": 190, "loss": 0.0013, "learning_rate": 6.1322604944307e-07, "epoch": 3.8451612903225807, "percentage": 78.42, "elapsed_time": "0:27:20", "remaining_time": "0:07:31", "throughput": "620.61", "total_tokens": 1018368}
+{"current_steps": 150, "total_steps": 190, "loss": 0.0003, "learning_rate": 5.848888922025553e-07, "epoch": 3.870967741935484, "percentage": 78.95, "elapsed_time": "0:27:31", "remaining_time": "0:07:20", "throughput": "620.76", "total_tokens": 1025424}
+{"current_steps": 151, "total_steps": 190, "loss": 0.0026, "learning_rate": 5.571350963575728e-07, "epoch": 3.896774193548387, "percentage": 79.47, "elapsed_time": "0:27:42", "remaining_time": "0:07:09", "throughput": "620.61", "total_tokens": 1032016}
+{"current_steps": 152, "total_steps": 190, "loss": 0.0097, "learning_rate": 5.299731159831953e-07, "epoch": 3.9225806451612906, "percentage": 80.0, "elapsed_time": "0:27:53", "remaining_time": "0:06:58", "throughput": "620.67", "total_tokens": 1038928}
+{"current_steps": 153, "total_steps": 190, "loss": 0.0047, "learning_rate": 5.034112248817685e-07, "epoch": 3.9483870967741934, "percentage": 80.53, "elapsed_time": "0:28:04", "remaining_time": "0:06:47", "throughput": "620.62", "total_tokens": 1045664}
+{"current_steps": 154, "total_steps": 190, "loss": 0.0081, "learning_rate": 4.774575140626317e-07, "epoch": 3.9741935483870967, "percentage": 81.05, "elapsed_time": "0:28:15", "remaining_time": "0:06:36", "throughput": "620.72", "total_tokens": 1052640}
+{"current_steps": 155, "total_steps": 190, "loss": 0.0018, "learning_rate": 4.5211988927752026e-07, "epoch": 4.0, "percentage": 81.58, "elapsed_time": "0:28:26", "remaining_time": "0:06:25", "throughput": "620.95", "total_tokens": 1059840}
+{"current_steps": 156, "total_steps": 190, "loss": 0.0053, "learning_rate": 4.27406068612396e-07, "epoch": 4.025806451612903, "percentage": 82.11, "elapsed_time": "0:28:37", "remaining_time": "0:06:14", "throughput": "620.97", "total_tokens": 1066704}
+{"current_steps": 157, "total_steps": 190, "loss": 0.0005, "learning_rate": 4.033235801364402e-07, "epoch": 4.051612903225807, "percentage": 82.63, "elapsed_time": "0:28:48", "remaining_time": "0:06:03", "throughput": "621.01", "total_tokens": 1073600}
+{"current_steps": 158, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.798797596089351e-07, "epoch": 4.077419354838709, "percentage": 83.16, "elapsed_time": "0:28:59", "remaining_time": "0:05:52", "throughput": "620.88", "total_tokens": 1080208}
+{"current_steps": 159, "total_steps": 190, "loss": 0.0018, "learning_rate": 3.5708174824471947e-07, "epoch": 4.103225806451613, "percentage": 83.68, "elapsed_time": "0:29:10", "remaining_time": "0:05:41", "throughput": "620.79", "total_tokens": 1086880}
+{"current_steps": 160, "total_steps": 190, "loss": 0.001, "learning_rate": 3.3493649053890325e-07, "epoch": 4.129032258064516, "percentage": 84.21, "elapsed_time": "0:29:21", "remaining_time": "0:05:30", "throughput": "620.68", "total_tokens": 1093520}
+{"current_steps": 161, "total_steps": 190, "loss": 0.0001, "learning_rate": 3.134507321515107e-07, "epoch": 4.15483870967742, "percentage": 84.74, "elapsed_time": "0:29:32", "remaining_time": "0:05:19", "throughput": "620.79", "total_tokens": 1100528}
+{"current_steps": 162, "total_steps": 190, "loss": 0.0012, "learning_rate": 2.9263101785268253e-07, "epoch": 4.180645161290323, "percentage": 85.26, "elapsed_time": "0:29:43", "remaining_time": "0:05:08", "throughput": "620.89", "total_tokens": 1107520}
+{"current_steps": 163, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.7248368952908055e-07, "epoch": 4.2064516129032254, "percentage": 85.79, "elapsed_time": "0:29:54", "remaining_time": "0:04:57", "throughput": "620.82", "total_tokens": 1114208}
+{"current_steps": 164, "total_steps": 190, "loss": 0.0002, "learning_rate": 2.53014884252083e-07, "epoch": 4.232258064516129, "percentage": 86.32, "elapsed_time": "0:30:05", "remaining_time": "0:04:46", "throughput": "620.76", "total_tokens": 1120912}
+{"current_steps": 165, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.3423053240837518e-07, "epoch": 4.258064516129032, "percentage": 86.84, "elapsed_time": "0:30:16", "remaining_time": "0:04:35", "throughput": "620.84", "total_tokens": 1127872}
+{"current_steps": 166, "total_steps": 190, "loss": 0.0003, "learning_rate": 2.1613635589349756e-07, "epoch": 4.283870967741936, "percentage": 87.37, "elapsed_time": "0:30:27", "remaining_time": "0:04:24", "throughput": "620.65", "total_tokens": 1134352}
+{"current_steps": 167, "total_steps": 190, "loss": 0.0001, "learning_rate": 1.9873786636889908e-07, "epoch": 4.309677419354839, "percentage": 87.89, "elapsed_time": "0:30:38", "remaining_time": "0:04:13", "throughput": "620.79", "total_tokens": 1141440}
+{"current_steps": 168, "total_steps": 190, "loss": 0.0007, "learning_rate": 1.8204036358303173e-07, "epoch": 4.335483870967742, "percentage": 88.42, "elapsed_time": "0:30:49", "remaining_time": "0:04:02", "throughput": "620.70", "total_tokens": 1148096}
+{"current_steps": 169, "total_steps": 190, "loss": 0.0003, "learning_rate": 1.6604893375699594e-07, "epoch": 4.361290322580645, "percentage": 88.95, "elapsed_time": "0:31:00", "remaining_time": "0:03:51", "throughput": "620.56", "total_tokens": 1154672}
+{"current_steps": 170, "total_steps": 190, "loss": 0.0001, "learning_rate": 1.507684480352292e-07, "epoch": 4.387096774193548, "percentage": 89.47, "elapsed_time": "0:31:11", "remaining_time": "0:03:40", "throughput": "620.70", "total_tokens": 1161744}
+{"current_steps": 171, "total_steps": 190, "loss": 0.0008, "learning_rate": 1.362035610017079e-07, "epoch": 4.412903225806452, "percentage": 90.0, "elapsed_time": "0:31:22", "remaining_time": "0:03:29", "throughput": "620.90", "total_tokens": 1168944}
+{"current_steps": 172, "total_steps": 190, "loss": 0.0001, "learning_rate": 1.223587092621162e-07, "epoch": 4.438709677419355, "percentage": 90.53, "elapsed_time": "0:31:33", "remaining_time": "0:03:18", "throughput": "620.92", "total_tokens": 1175792}
+{"current_steps": 173, "total_steps": 190, "loss": 0.0004, "learning_rate": 1.0923811009241142e-07, "epoch": 4.464516129032258, "percentage": 91.05, "elapsed_time": "0:31:44", "remaining_time": "0:03:07", "throughput": "621.02", "total_tokens": 1182800}
+{"current_steps": 174, "total_steps": 190, "loss": 0.0001, "learning_rate": 9.684576015420277e-08, "epoch": 4.490322580645161, "percentage": 91.58, "elapsed_time": "0:31:55", "remaining_time": "0:02:56", "throughput": "620.99", "total_tokens": 1189568}
+{"current_steps": 175, "total_steps": 190, "loss": 0.0005, "learning_rate": 8.518543427732951e-08, "epoch": 4.516129032258064, "percentage": 92.11, "elapsed_time": "0:32:06", "remaining_time": "0:02:45", "throughput": "621.03", "total_tokens": 1196480}
+{"current_steps": 176, "total_steps": 190, "loss": 0.0076, "learning_rate": 7.426068431000883e-08, "epoch": 4.541935483870968, "percentage": 92.63, "elapsed_time": "0:32:17", "remaining_time": "0:02:34", "throughput": "621.25", "total_tokens": 1203728}
+{"current_steps": 177, "total_steps": 190, "loss": 0.0004, "learning_rate": 6.407483803691216e-08, "epoch": 4.567741935483871, "percentage": 93.16, "elapsed_time": "0:32:28", "remaining_time": "0:02:23", "throughput": "621.12", "total_tokens": 1210320}
+{"current_steps": 178, "total_steps": 190, "loss": 0.004, "learning_rate": 5.463099816548578e-08, "epoch": 4.593548387096774, "percentage": 93.68, "elapsed_time": "0:32:39", "remaining_time": "0:02:12", "throughput": "621.16", "total_tokens": 1217248}
+{"current_steps": 179, "total_steps": 190, "loss": 0.0001, "learning_rate": 4.593204138084006e-08, "epoch": 4.619354838709677, "percentage": 94.21, "elapsed_time": "0:32:50", "remaining_time": "0:02:01", "throughput": "621.22", "total_tokens": 1224192}
+{"current_steps": 180, "total_steps": 190, "loss": 0.0005, "learning_rate": 3.798061746947995e-08, "epoch": 4.645161290322581, "percentage": 94.74, "elapsed_time": "0:33:01", "remaining_time": "0:01:50", "throughput": "621.17", "total_tokens": 1230896}
+{"current_steps": 181, "total_steps": 190, "loss": 0.0002, "learning_rate": 3.077914851215585e-08, "epoch": 4.670967741935484, "percentage": 95.26, "elapsed_time": "0:33:12", "remaining_time": "0:01:39", "throughput": "621.23", "total_tokens": 1237840}
+{"current_steps": 182, "total_steps": 190, "loss": 0.0001, "learning_rate": 2.4329828146074096e-08, "epoch": 4.6967741935483875, "percentage": 95.79, "elapsed_time": "0:33:23", "remaining_time": "0:01:28", "throughput": "621.25", "total_tokens": 1244688}
+{"current_steps": 183, "total_steps": 190, "loss": 0.0001, "learning_rate": 1.8634620896695044e-08, "epoch": 4.72258064516129, "percentage": 96.32, "elapsed_time": "0:33:34", "remaining_time": "0:01:17", "throughput": "621.28", "total_tokens": 1251584}
+{"current_steps": 184, "total_steps": 190, "loss": 0.0001, "learning_rate": 1.3695261579316776e-08, "epoch": 4.748387096774193, "percentage": 96.84, "elapsed_time": "0:33:45", "remaining_time": "0:01:06", "throughput": "621.30", "total_tokens": 1258448}
+{"current_steps": 185, "total_steps": 190, "loss": 0.0167, "learning_rate": 9.513254770636138e-09, "epoch": 4.774193548387097, "percentage": 97.37, "elapsed_time": "0:33:56", "remaining_time": "0:00:55", "throughput": "621.38", "total_tokens": 1265440}
+{"current_steps": 186, "total_steps": 190, "loss": 0.0002, "learning_rate": 6.089874350439507e-09, "epoch": 4.8, "percentage": 97.89, "elapsed_time": "0:34:07", "remaining_time": "0:00:44", "throughput": "621.28", "total_tokens": 1272080}
+{"current_steps": 187, "total_steps": 190, "loss": 0.0003, "learning_rate": 3.4261631135654174e-09, "epoch": 4.825806451612904, "percentage": 98.42, "elapsed_time": "0:34:18", "remaining_time": "0:00:33", "throughput": "621.31", "total_tokens": 1278976}
+{"current_steps": 188, "total_steps": 190, "loss": 0.0024, "learning_rate": 1.5229324522605949e-09, "epoch": 4.851612903225806, "percentage": 98.95, "elapsed_time": "0:34:29", "remaining_time": "0:00:22", "throughput": "621.29", "total_tokens": 1285760}
+{"current_steps": 189, "total_steps": 190, "loss": 0.0005, "learning_rate": 3.8076210902182607e-10, "epoch": 4.877419354838709, "percentage": 99.47, "elapsed_time": "0:34:40", "remaining_time": "0:00:11", "throughput": "621.19", "total_tokens": 1292368}
+{"current_steps": 190, "total_steps": 190, "loss": 0.0017, "learning_rate": 0.0, "epoch": 4.903225806451613, "percentage": 100.0, "elapsed_time": "0:34:51", "remaining_time": "0:00:00", "throughput": "621.15", "total_tokens": 1299120}
+{"current_steps": 190, "total_steps": 190, "epoch": 4.903225806451613, "percentage": 100.0, "elapsed_time": "0:35:42", "remaining_time": "0:00:00", "throughput": "606.30", "total_tokens": 1299120}

trainer_state.json

@@ -0,0 +1,1563 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.903225806451613,
+  "eval_steps": 500,
+  "global_step": 190,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.025806451612903226,
+      "grad_norm": 276.5378112792969,
+      "learning_rate": 5.000000000000001e-07,
+      "loss": 8.2514,
+      "num_input_tokens_seen": 6848,
+      "step": 1
+    },
+    {
+      "epoch": 0.05161290322580645,
+      "grad_norm": 294.3509521484375,
+      "learning_rate": 1.0000000000000002e-06,
+      "loss": 8.2793,
+      "num_input_tokens_seen": 13776,
+      "step": 2
+    },
+    {
+      "epoch": 0.07741935483870968,
+      "grad_norm": 283.4303283691406,
+      "learning_rate": 1.5e-06,
+      "loss": 8.17,
+      "num_input_tokens_seen": 20672,
+      "step": 3
+    },
+    {
+      "epoch": 0.1032258064516129,
+      "grad_norm": 285.55255126953125,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 7.6197,
+      "num_input_tokens_seen": 27744,
+      "step": 4
+    },
+    {
+      "epoch": 0.12903225806451613,
+      "grad_norm": 271.7382507324219,
+      "learning_rate": 2.5e-06,
+      "loss": 6.9491,
+      "num_input_tokens_seen": 34624,
+      "step": 5
+    },
+    {
+      "epoch": 0.15483870967741936,
+      "grad_norm": 125.66632080078125,
+      "learning_rate": 3e-06,
+      "loss": 5.2054,
+      "num_input_tokens_seen": 41424,
+      "step": 6
+    },
+    {
+      "epoch": 0.18064516129032257,
+      "grad_norm": 113.50272369384766,
+      "learning_rate": 3.5e-06,
+      "loss": 4.8642,
+      "num_input_tokens_seen": 48304,
+      "step": 7
+    },
+    {
+      "epoch": 0.2064516129032258,
+      "grad_norm": 113.24736785888672,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 3.2874,
+      "num_input_tokens_seen": 55152,
+      "step": 8
+    },
+    {
+      "epoch": 0.23225806451612904,
+      "grad_norm": 117.07583618164062,
+      "learning_rate": 4.5e-06,
+      "loss": 2.631,
+      "num_input_tokens_seen": 61680,
+      "step": 9
+    },
+    {
+      "epoch": 0.25806451612903225,
+      "grad_norm": 113.77539825439453,
+      "learning_rate": 5e-06,
+      "loss": 0.6982,
+      "num_input_tokens_seen": 68480,
+      "step": 10
+    },
+    {
+      "epoch": 0.2838709677419355,
+      "grad_norm": 36.774471282958984,
+      "learning_rate": 4.9996192378909785e-06,
+      "loss": 0.3276,
+      "num_input_tokens_seen": 75296,
+      "step": 11
+    },
+    {
+      "epoch": 0.3096774193548387,
+      "grad_norm": 33.21468734741211,
+      "learning_rate": 4.99847706754774e-06,
+      "loss": 0.293,
+      "num_input_tokens_seen": 82288,
+      "step": 12
+    },
+    {
+      "epoch": 0.33548387096774196,
+      "grad_norm": 5.65634822845459,
+      "learning_rate": 4.9965738368864345e-06,
+      "loss": 0.2129,
+      "num_input_tokens_seen": 89088,
+      "step": 13
+    },
+    {
+      "epoch": 0.36129032258064514,
+      "grad_norm": 55.87510299682617,
+      "learning_rate": 4.993910125649561e-06,
+      "loss": 0.4712,
+      "num_input_tokens_seen": 95904,
+      "step": 14
+    },
+    {
+      "epoch": 0.3870967741935484,
+      "grad_norm": 31.815500259399414,
+      "learning_rate": 4.990486745229364e-06,
+      "loss": 0.235,
+      "num_input_tokens_seen": 102896,
+      "step": 15
+    },
+    {
+      "epoch": 0.4129032258064516,
+      "grad_norm": 10.451783180236816,
+      "learning_rate": 4.986304738420684e-06,
+      "loss": 0.202,
+      "num_input_tokens_seen": 109792,
+      "step": 16
+    },
+    {
+      "epoch": 0.43870967741935485,
+      "grad_norm": 21.014144897460938,
+      "learning_rate": 4.981365379103306e-06,
+      "loss": 0.1981,
+      "num_input_tokens_seen": 116528,
+      "step": 17
+    },
+    {
+      "epoch": 0.4645161290322581,
+      "grad_norm": 4.897561550140381,
+      "learning_rate": 4.975670171853926e-06,
+      "loss": 0.1517,
+      "num_input_tokens_seen": 123200,
+      "step": 18
+    },
+    {
+      "epoch": 0.49032258064516127,
+      "grad_norm": 37.270225524902344,
+      "learning_rate": 4.9692208514878445e-06,
+      "loss": 0.4335,
+      "num_input_tokens_seen": 130032,
+      "step": 19
+    },
+    {
+      "epoch": 0.5161290322580645,
+      "grad_norm": 31.595401763916016,
+      "learning_rate": 4.962019382530521e-06,
+      "loss": 0.3609,
+      "num_input_tokens_seen": 136832,
+      "step": 20
+    },
+    {
+      "epoch": 0.5419354838709678,
+      "grad_norm": 8.859624862670898,
+      "learning_rate": 4.9540679586191605e-06,
+      "loss": 0.1708,
+      "num_input_tokens_seen": 143904,
+      "step": 21
+    },
+    {
+      "epoch": 0.567741935483871,
+      "grad_norm": 21.666885375976562,
+      "learning_rate": 4.9453690018345144e-06,
+      "loss": 0.2277,
+      "num_input_tokens_seen": 150496,
+      "step": 22
+    },
+    {
+      "epoch": 0.5935483870967742,
+      "grad_norm": 29.53874969482422,
+      "learning_rate": 4.935925161963089e-06,
+      "loss": 0.3437,
+      "num_input_tokens_seen": 157360,
+      "step": 23
+    },
+    {
+      "epoch": 0.6193548387096774,
+      "grad_norm": 17.89641761779785,
+      "learning_rate": 4.925739315689991e-06,
+      "loss": 0.2229,
+      "num_input_tokens_seen": 164464,
+      "step": 24
+    },
+    {
+      "epoch": 0.6451612903225806,
+      "grad_norm": 2.7967522144317627,
+      "learning_rate": 4.914814565722671e-06,
+      "loss": 0.1242,
+      "num_input_tokens_seen": 170960,
+      "step": 25
+    },
+    {
+      "epoch": 0.6709677419354839,
+      "grad_norm": 15.343024253845215,
+      "learning_rate": 4.903154239845798e-06,
+      "loss": 0.2117,
+      "num_input_tokens_seen": 177808,
+      "step": 26
+    },
+    {
+      "epoch": 0.6967741935483871,
+      "grad_norm": 20.621328353881836,
+      "learning_rate": 4.890761889907589e-06,
+      "loss": 0.2706,
+      "num_input_tokens_seen": 184848,
+      "step": 27
+    },
+    {
+      "epoch": 0.7225806451612903,
+      "grad_norm": 14.205361366271973,
+      "learning_rate": 4.8776412907378845e-06,
+      "loss": 0.2084,
+      "num_input_tokens_seen": 191536,
+      "step": 28
+    },
+    {
+      "epoch": 0.7483870967741936,
+      "grad_norm": 4.756993770599365,
+      "learning_rate": 4.863796438998293e-06,
+      "loss": 0.0981,
+      "num_input_tokens_seen": 198368,
+      "step": 29
+    },
+    {
+      "epoch": 0.7741935483870968,
+      "grad_norm": 13.791275024414062,
+      "learning_rate": 4.849231551964771e-06,
+      "loss": 0.16,
+      "num_input_tokens_seen": 205200,
+      "step": 30
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 11.285390853881836,
+      "learning_rate": 4.833951066243004e-06,
+      "loss": 0.1614,
+      "num_input_tokens_seen": 211760,
+      "step": 31
+    },
+    {
+      "epoch": 0.8258064516129032,
+      "grad_norm": 13.351017951965332,
+      "learning_rate": 4.817959636416969e-06,
+      "loss": 0.1742,
+      "num_input_tokens_seen": 218448,
+      "step": 32
+    },
+    {
+      "epoch": 0.8516129032258064,
+      "grad_norm": 6.028775691986084,
+      "learning_rate": 4.801262133631101e-06,
+      "loss": 0.1107,
+      "num_input_tokens_seen": 225408,
+      "step": 33
+    },
+    {
+      "epoch": 0.8774193548387097,
+      "grad_norm": 2.5520031452178955,
+      "learning_rate": 4.783863644106502e-06,
+      "loss": 0.0822,
+      "num_input_tokens_seen": 232032,
+      "step": 34
+    },
+    {
+      "epoch": 0.9032258064516129,
+      "grad_norm": 13.276092529296875,
+      "learning_rate": 4.765769467591626e-06,
+      "loss": 0.1873,
+      "num_input_tokens_seen": 238672,
+      "step": 35
+    },
+    {
+      "epoch": 0.9290322580645162,
+      "grad_norm": 15.978204727172852,
+      "learning_rate": 4.746985115747918e-06,
+      "loss": 0.2375,
+      "num_input_tokens_seen": 245456,
+      "step": 36
+    },
+    {
+      "epoch": 0.9548387096774194,
+      "grad_norm": 16.432758331298828,
+      "learning_rate": 4.72751631047092e-06,
+      "loss": 0.2667,
+      "num_input_tokens_seen": 252576,
+      "step": 37
+    },
+    {
+      "epoch": 0.9806451612903225,
+      "grad_norm": 12.175621032714844,
+      "learning_rate": 4.707368982147318e-06,
+      "loss": 0.1547,
+      "num_input_tokens_seen": 259536,
+      "step": 38
+    },
+    {
+      "epoch": 1.0064516129032257,
+      "grad_norm": 9.780308723449707,
+      "learning_rate": 4.68654926784849e-06,
+      "loss": 0.1662,
+      "num_input_tokens_seen": 266560,
+      "step": 39
+    },
+    {
+      "epoch": 1.032258064516129,
+      "grad_norm": 3.619741678237915,
+      "learning_rate": 4.665063509461098e-06,
+      "loss": 0.0808,
+      "num_input_tokens_seen": 273248,
+      "step": 40
+    },
+    {
+      "epoch": 1.0580645161290323,
+      "grad_norm": 5.894082546234131,
+      "learning_rate": 4.642918251755281e-06,
+      "loss": 0.0884,
+      "num_input_tokens_seen": 279888,
+      "step": 41
+    },
+    {
+      "epoch": 1.0838709677419356,
+      "grad_norm": 5.443681240081787,
+      "learning_rate": 4.620120240391065e-06,
+      "loss": 0.0883,
+      "num_input_tokens_seen": 286944,
+      "step": 42
+    },
+    {
+      "epoch": 1.1096774193548387,
+      "grad_norm": 3.3760452270507812,
+      "learning_rate": 4.596676419863561e-06,
+      "loss": 0.0562,
+      "num_input_tokens_seen": 293792,
+      "step": 43
+    },
+    {
+      "epoch": 1.135483870967742,
+      "grad_norm": 2.7039427757263184,
+      "learning_rate": 4.572593931387604e-06,
+      "loss": 0.0856,
+      "num_input_tokens_seen": 300512,
+      "step": 44
+    },
+    {
+      "epoch": 1.1612903225806452,
+      "grad_norm": 4.186523914337158,
+      "learning_rate": 4.54788011072248e-06,
+      "loss": 0.0612,
+      "num_input_tokens_seen": 307264,
+      "step": 45
+    },
+    {
+      "epoch": 1.1870967741935483,
+      "grad_norm": 4.6027374267578125,
+      "learning_rate": 4.522542485937369e-06,
+      "loss": 0.0944,
+      "num_input_tokens_seen": 314112,
+      "step": 46
+    },
+    {
+      "epoch": 1.2129032258064516,
+      "grad_norm": 4.636350154876709,
+      "learning_rate": 4.496588775118232e-06,
+      "loss": 0.0624,
+      "num_input_tokens_seen": 320864,
+      "step": 47
+    },
+    {
+      "epoch": 1.238709677419355,
+      "grad_norm": 3.0458312034606934,
+      "learning_rate": 4.470026884016805e-06,
+      "loss": 0.0363,
+      "num_input_tokens_seen": 327648,
+      "step": 48
+    },
+    {
+      "epoch": 1.2645161290322582,
+      "grad_norm": 4.572902679443359,
+      "learning_rate": 4.442864903642428e-06,
+      "loss": 0.1039,
+      "num_input_tokens_seen": 334432,
+      "step": 49
+    },
+    {
+      "epoch": 1.2903225806451613,
+      "grad_norm": 2.629575252532959,
+      "learning_rate": 4.415111107797445e-06,
+      "loss": 0.0488,
+      "num_input_tokens_seen": 341248,
+      "step": 50
+    },
+    {
+      "epoch": 1.3161290322580645,
+      "grad_norm": 2.7692055702209473,
+      "learning_rate": 4.386773950556931e-06,
+      "loss": 0.0613,
+      "num_input_tokens_seen": 348192,
+      "step": 51
+    },
+    {
+      "epoch": 1.3419354838709676,
+      "grad_norm": 2.706336498260498,
+      "learning_rate": 4.357862063693486e-06,
+      "loss": 0.07,
+      "num_input_tokens_seen": 354816,
+      "step": 52
+    },
+    {
+      "epoch": 1.367741935483871,
+      "grad_norm": 1.679677963256836,
+      "learning_rate": 4.328384254047927e-06,
+      "loss": 0.0463,
+      "num_input_tokens_seen": 361600,
+      "step": 53
+    },
+    {
+      "epoch": 1.3935483870967742,
+      "grad_norm": 2.623893976211548,
+      "learning_rate": 4.2983495008466285e-06,
+      "loss": 0.0671,
+      "num_input_tokens_seen": 368464,
+      "step": 54
+    },
+    {
+      "epoch": 1.4193548387096775,
+      "grad_norm": 1.7314574718475342,
+      "learning_rate": 4.267766952966369e-06,
+      "loss": 0.0428,
+      "num_input_tokens_seen": 375440,
+      "step": 55
+    },
+    {
+      "epoch": 1.4451612903225808,
+      "grad_norm": 5.240540981292725,
+      "learning_rate": 4.236645926147493e-06,
+      "loss": 0.0678,
+      "num_input_tokens_seen": 382240,
+      "step": 56
+    },
+    {
+      "epoch": 1.4709677419354839,
+      "grad_norm": 1.9007383584976196,
+      "learning_rate": 4.204995900156247e-06,
+      "loss": 0.0476,
+      "num_input_tokens_seen": 389008,
+      "step": 57
+    },
+    {
+      "epoch": 1.4967741935483871,
+      "grad_norm": 3.100925922393799,
+      "learning_rate": 4.172826515897146e-06,
+      "loss": 0.0442,
+      "num_input_tokens_seen": 396080,
+      "step": 58
+    },
+    {
+      "epoch": 1.5225806451612902,
+      "grad_norm": 2.798711061477661,
+      "learning_rate": 4.140147572476269e-06,
+      "loss": 0.0336,
+      "num_input_tokens_seen": 403056,
+      "step": 59
+    },
+    {
+      "epoch": 1.5483870967741935,
+      "grad_norm": 2.20042085647583,
+      "learning_rate": 4.106969024216348e-06,
+      "loss": 0.046,
+      "num_input_tokens_seen": 409664,
+      "step": 60
+    },
+    {
+      "epoch": 1.5741935483870968,
+      "grad_norm": 4.055737495422363,
+      "learning_rate": 4.073300977624594e-06,
+      "loss": 0.0416,
+      "num_input_tokens_seen": 416272,
+      "step": 61
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 3.15283465385437,
+      "learning_rate": 4.039153688314146e-06,
+      "loss": 0.0649,
+      "num_input_tokens_seen": 423360,
+      "step": 62
+    },
+    {
+      "epoch": 1.6258064516129034,
+      "grad_norm": 4.418552398681641,
+      "learning_rate": 4.0045375578801216e-06,
+      "loss": 0.0591,
+      "num_input_tokens_seen": 430304,
+      "step": 63
+    },
+    {
+      "epoch": 1.6516129032258065,
+      "grad_norm": 3.960134506225586,
+      "learning_rate": 3.969463130731183e-06,
+      "loss": 0.0318,
+      "num_input_tokens_seen": 436960,
+      "step": 64
+    },
+    {
+      "epoch": 1.6774193548387095,
+      "grad_norm": 3.1769161224365234,
+      "learning_rate": 3.933941090877615e-06,
+      "loss": 0.0462,
+      "num_input_tokens_seen": 443792,
+      "step": 65
+    },
+    {
+      "epoch": 1.7032258064516128,
+      "grad_norm": 4.235939025878906,
+      "learning_rate": 3.897982258676867e-06,
+      "loss": 0.0465,
+      "num_input_tokens_seen": 450672,
+      "step": 66
+    },
+    {
+      "epoch": 1.729032258064516,
+      "grad_norm": 2.7547543048858643,
+      "learning_rate": 3.861597587537568e-06,
+      "loss": 0.0316,
+      "num_input_tokens_seen": 457616,
+      "step": 67
+    },
+    {
+      "epoch": 1.7548387096774194,
+      "grad_norm": 4.103342533111572,
+      "learning_rate": 3.824798160583012e-06,
+      "loss": 0.1,
+      "num_input_tokens_seen": 464592,
+      "step": 68
+    },
+    {
+      "epoch": 1.7806451612903227,
+      "grad_norm": 3.525808811187744,
+      "learning_rate": 3.787595187275136e-06,
+      "loss": 0.0711,
+      "num_input_tokens_seen": 471328,
+      "step": 69
+    },
+    {
+      "epoch": 1.8064516129032258,
+      "grad_norm": 2.9660446643829346,
+      "learning_rate": 3.7500000000000005e-06,
+      "loss": 0.0494,
+      "num_input_tokens_seen": 478480,
+      "step": 70
+    },
+    {
+      "epoch": 1.832258064516129,
+      "grad_norm": 3.661849021911621,
+      "learning_rate": 3.7120240506158433e-06,
+      "loss": 0.0618,
+      "num_input_tokens_seen": 485264,
+      "step": 71
+    },
+    {
+      "epoch": 1.8580645161290321,
+      "grad_norm": 1.8227900266647339,
+      "learning_rate": 3.6736789069647273e-06,
+      "loss": 0.0511,
+      "num_input_tokens_seen": 492064,
+      "step": 72
+    },
+    {
+      "epoch": 1.8838709677419354,
+      "grad_norm": 2.1322553157806396,
+      "learning_rate": 3.634976249348867e-06,
+      "loss": 0.0464,
+      "num_input_tokens_seen": 498896,
+      "step": 73
+    },
+    {
+      "epoch": 1.9096774193548387,
+      "grad_norm": 1.8635531663894653,
+      "learning_rate": 3.595927866972694e-06,
+      "loss": 0.0331,
+      "num_input_tokens_seen": 506016,
+      "step": 74
+    },
+    {
+      "epoch": 1.935483870967742,
+      "grad_norm": 2.9526419639587402,
+      "learning_rate": 3.556545654351749e-06,
+      "loss": 0.0706,
+      "num_input_tokens_seen": 512848,
+      "step": 75
+    },
+    {
+      "epoch": 1.9612903225806453,
+      "grad_norm": 2.8351802825927734,
+      "learning_rate": 3.516841607689501e-06,
+      "loss": 0.0442,
+      "num_input_tokens_seen": 519744,
+      "step": 76
+    },
+    {
+      "epoch": 1.9870967741935484,
+      "grad_norm": 3.1449944972991943,
+      "learning_rate": 3.476827821223184e-06,
+      "loss": 0.042,
+      "num_input_tokens_seen": 526352,
+      "step": 77
+    },
+    {
+      "epoch": 2.0129032258064514,
+      "grad_norm": 1.6314219236373901,
+      "learning_rate": 3.436516483539781e-06,
+      "loss": 0.021,
+      "num_input_tokens_seen": 533136,
+      "step": 78
+    },
+    {
+      "epoch": 2.0387096774193547,
+      "grad_norm": 1.6224279403686523,
+      "learning_rate": 3.39591987386325e-06,
+      "loss": 0.0094,
+      "num_input_tokens_seen": 540016,
+      "step": 79
+    },
+    {
+      "epoch": 2.064516129032258,
+      "grad_norm": 0.3502344787120819,
+      "learning_rate": 3.3550503583141726e-06,
+      "loss": 0.0021,
+      "num_input_tokens_seen": 547152,
+      "step": 80
+    },
+    {
+      "epoch": 2.0903225806451613,
+      "grad_norm": 1.7473047971725464,
+      "learning_rate": 3.313920386142892e-06,
+      "loss": 0.0146,
+      "num_input_tokens_seen": 553888,
+      "step": 81
+    },
+    {
+      "epoch": 2.1161290322580646,
+      "grad_norm": 2.363898277282715,
+      "learning_rate": 3.272542485937369e-06,
+      "loss": 0.0237,
+      "num_input_tokens_seen": 561024,
+      "step": 82
+    },
+    {
+      "epoch": 2.141935483870968,
+      "grad_norm": 1.377082109451294,
+      "learning_rate": 3.230929261806842e-06,
+      "loss": 0.0031,
+      "num_input_tokens_seen": 567936,
+      "step": 83
+    },
+    {
+      "epoch": 2.167741935483871,
+      "grad_norm": 0.6467981934547424,
+      "learning_rate": 3.189093389542498e-06,
+      "loss": 0.0034,
+      "num_input_tokens_seen": 574736,
+      "step": 84
+    },
+    {
+      "epoch": 2.193548387096774,
+      "grad_norm": 0.9653486609458923,
+      "learning_rate": 3.147047612756302e-06,
+      "loss": 0.0045,
+      "num_input_tokens_seen": 581680,
+      "step": 85
+    },
+    {
+      "epoch": 2.2193548387096773,
+      "grad_norm": 0.7019590735435486,
+      "learning_rate": 3.1048047389991693e-06,
+      "loss": 0.0031,
+      "num_input_tokens_seen": 588464,
+      "step": 86
+    },
+    {
+      "epoch": 2.2451612903225806,
+      "grad_norm": 2.078730821609497,
+      "learning_rate": 3.062377635859663e-06,
+      "loss": 0.0341,
+      "num_input_tokens_seen": 595312,
+      "step": 87
+    },
+    {
+      "epoch": 2.270967741935484,
+      "grad_norm": 4.608306407928467,
+      "learning_rate": 3.019779227044398e-06,
+      "loss": 0.0095,
+      "num_input_tokens_seen": 602080,
+      "step": 88
+    },
+    {
+      "epoch": 2.296774193548387,
+      "grad_norm": 3.8219709396362305,
+      "learning_rate": 2.9770224884413625e-06,
+      "loss": 0.0459,
+      "num_input_tokens_seen": 609168,
+      "step": 89
+    },
+    {
+      "epoch": 2.3225806451612905,
+      "grad_norm": 2.868053674697876,
+      "learning_rate": 2.9341204441673267e-06,
+      "loss": 0.0104,
+      "num_input_tokens_seen": 616000,
+      "step": 90
+    },
+    {
+      "epoch": 2.3483870967741938,
+      "grad_norm": 2.516240119934082,
+      "learning_rate": 2.8910861626005774e-06,
+      "loss": 0.0201,
+      "num_input_tokens_seen": 622544,
+      "step": 91
+    },
+    {
+      "epoch": 2.3741935483870966,
+      "grad_norm": 0.5750585794448853,
+      "learning_rate": 2.847932752400164e-06,
+      "loss": 0.0021,
+      "num_input_tokens_seen": 629600,
+      "step": 92
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 2.9817891120910645,
+      "learning_rate": 2.804673358512869e-06,
+      "loss": 0.043,
+      "num_input_tokens_seen": 636320,
+      "step": 93
+    },
+    {
+      "epoch": 2.425806451612903,
+      "grad_norm": 1.8277342319488525,
+      "learning_rate": 2.761321158169134e-06,
+      "loss": 0.0207,
+      "num_input_tokens_seen": 643136,
+      "step": 94
+    },
+    {
+      "epoch": 2.4516129032258065,
+      "grad_norm": 4.615224361419678,
+      "learning_rate": 2.717889356869146e-06,
+      "loss": 0.0148,
+      "num_input_tokens_seen": 650048,
+      "step": 95
+    },
+    {
+      "epoch": 2.47741935483871,
+      "grad_norm": 0.9854787588119507,
+      "learning_rate": 2.6743911843603134e-06,
+      "loss": 0.004,
+      "num_input_tokens_seen": 656736,
+      "step": 96
+    },
+    {
+      "epoch": 2.5032258064516126,
+      "grad_norm": 1.4292289018630981,
+      "learning_rate": 2.6308398906073603e-06,
+      "loss": 0.0131,
+      "num_input_tokens_seen": 663552,
+      "step": 97
+    },
+    {
+      "epoch": 2.5290322580645164,
+      "grad_norm": 4.594967842102051,
+      "learning_rate": 2.587248741756253e-06,
+      "loss": 0.0455,
+      "num_input_tokens_seen": 670224,
+      "step": 98
+    },
+    {
+      "epoch": 2.554838709677419,
+      "grad_norm": 0.8579672574996948,
+      "learning_rate": 2.543631016093209e-06,
+      "loss": 0.0031,
+      "num_input_tokens_seen": 676960,
+      "step": 99
+    },
+    {
+      "epoch": 2.5806451612903225,
+      "grad_norm": 2.075626850128174,
+      "learning_rate": 2.5e-06,
+      "loss": 0.0099,
+      "num_input_tokens_seen": 684016,
+      "step": 100
+    },
+    {
+      "epoch": 2.606451612903226,
+      "grad_norm": 3.9176177978515625,
+      "learning_rate": 2.4563689839067913e-06,
+      "loss": 0.0797,
+      "num_input_tokens_seen": 690832,
+      "step": 101
+    },
+    {
+      "epoch": 2.632258064516129,
+      "grad_norm": 1.6330746412277222,
+      "learning_rate": 2.4127512582437486e-06,
+      "loss": 0.0059,
+      "num_input_tokens_seen": 697808,
+      "step": 102
+    },
+    {
+      "epoch": 2.6580645161290324,
+      "grad_norm": 2.173410177230835,
+      "learning_rate": 2.3691601093926406e-06,
+      "loss": 0.0438,
+      "num_input_tokens_seen": 704352,
+      "step": 103
+    },
+    {
+      "epoch": 2.6838709677419352,
+      "grad_norm": 1.6522724628448486,
+      "learning_rate": 2.325608815639687e-06,
+      "loss": 0.0149,
+      "num_input_tokens_seen": 711456,
+      "step": 104
+    },
+    {
+      "epoch": 2.709677419354839,
+      "grad_norm": 0.7776157855987549,
+      "learning_rate": 2.2821106431308546e-06,
+      "loss": 0.0126,
+      "num_input_tokens_seen": 718224,
+      "step": 105
+    },
+    {
+      "epoch": 2.735483870967742,
+      "grad_norm": 2.2912323474884033,
+      "learning_rate": 2.238678841830867e-06,
+      "loss": 0.0255,
+      "num_input_tokens_seen": 724928,
+      "step": 106
+    },
+    {
+      "epoch": 2.761290322580645,
+      "grad_norm": 0.5118953585624695,
+      "learning_rate": 2.195326641487132e-06,
+      "loss": 0.0048,
+      "num_input_tokens_seen": 731616,
+      "step": 107
+    },
+    {
+      "epoch": 2.7870967741935484,
+      "grad_norm": 0.9067234396934509,
+      "learning_rate": 2.1520672475998374e-06,
+      "loss": 0.0142,
+      "num_input_tokens_seen": 738272,
+      "step": 108
+    },
+    {
+      "epoch": 2.8129032258064517,
+      "grad_norm": 1.6885769367218018,
+      "learning_rate": 2.1089138373994226e-06,
+      "loss": 0.0193,
+      "num_input_tokens_seen": 745104,
+      "step": 109
+    },
+    {
+      "epoch": 2.838709677419355,
+      "grad_norm": 0.6433587074279785,
+      "learning_rate": 2.0658795558326745e-06,
+      "loss": 0.0055,
+      "num_input_tokens_seen": 752016,
+      "step": 110
+    },
+    {
+      "epoch": 2.864516129032258,
+      "grad_norm": 1.0263943672180176,
+      "learning_rate": 2.022977511558638e-06,
+      "loss": 0.0144,
+      "num_input_tokens_seen": 758720,
+      "step": 111
+    },
+    {
+      "epoch": 2.8903225806451616,
+      "grad_norm": 1.3137305974960327,
+      "learning_rate": 1.9802207729556023e-06,
+      "loss": 0.0272,
+      "num_input_tokens_seen": 765520,
+      "step": 112
+    },
+    {
+      "epoch": 2.9161290322580644,
+      "grad_norm": 2.692868947982788,
+      "learning_rate": 1.937622364140338e-06,
+      "loss": 0.0101,
+      "num_input_tokens_seen": 772288,
+      "step": 113
+    },
+    {
+      "epoch": 2.9419354838709677,
+      "grad_norm": 2.19588565826416,
+      "learning_rate": 1.895195261000831e-06,
+      "loss": 0.0109,
+      "num_input_tokens_seen": 779520,
+      "step": 114
+    },
+    {
+      "epoch": 2.967741935483871,
+      "grad_norm": 2.681143283843994,
+      "learning_rate": 1.852952387243698e-06,
+      "loss": 0.018,
+      "num_input_tokens_seen": 786400,
+      "step": 115
+    },
+    {
+      "epoch": 2.9935483870967743,
+      "grad_norm": 1.9662718772888184,
+      "learning_rate": 1.8109066104575023e-06,
+      "loss": 0.0141,
+      "num_input_tokens_seen": 793008,
+      "step": 116
+    },
+    {
+      "epoch": 3.0193548387096776,
+      "grad_norm": 1.1327028274536133,
+      "learning_rate": 1.7690707381931585e-06,
+      "loss": 0.0057,
+      "num_input_tokens_seen": 799808,
+      "step": 117
+    },
+    {
+      "epoch": 3.0451612903225804,
+      "grad_norm": 0.593132495880127,
+      "learning_rate": 1.7274575140626318e-06,
+      "loss": 0.0063,
+      "num_input_tokens_seen": 806576,
+      "step": 118
+    },
+    {
+      "epoch": 3.0709677419354837,
+      "grad_norm": 1.123922348022461,
+      "learning_rate": 1.686079613857109e-06,
+      "loss": 0.0138,
+      "num_input_tokens_seen": 813312,
+      "step": 119
+    },
+    {
+      "epoch": 3.096774193548387,
+      "grad_norm": 0.14445500075817108,
+      "learning_rate": 1.6449496416858285e-06,
+      "loss": 0.0011,
+      "num_input_tokens_seen": 819728,
+      "step": 120
+    },
+    {
+      "epoch": 3.1225806451612903,
+      "grad_norm": 0.09276027232408524,
+      "learning_rate": 1.6040801261367494e-06,
+      "loss": 0.0006,
+      "num_input_tokens_seen": 826496,
+      "step": 121
+    },
+    {
+      "epoch": 3.1483870967741936,
+      "grad_norm": 0.6197965741157532,
+      "learning_rate": 1.56348351646022e-06,
+      "loss": 0.0055,
+      "num_input_tokens_seen": 833360,
+      "step": 122
+    },
+    {
+      "epoch": 3.174193548387097,
+      "grad_norm": 0.1872132420539856,
+      "learning_rate": 1.5231721787768162e-06,
+      "loss": 0.0011,
+      "num_input_tokens_seen": 840080,
+      "step": 123
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 2.89666748046875,
+      "learning_rate": 1.4831583923105e-06,
+      "loss": 0.0173,
+      "num_input_tokens_seen": 847104,
+      "step": 124
+    },
+    {
+      "epoch": 3.225806451612903,
+      "grad_norm": 0.8067394495010376,
+      "learning_rate": 1.443454345648252e-06,
+      "loss": 0.0027,
+      "num_input_tokens_seen": 853712,
+      "step": 125
+    },
+    {
+      "epoch": 3.2516129032258063,
+      "grad_norm": 0.5419718623161316,
+      "learning_rate": 1.4040721330273063e-06,
+      "loss": 0.0029,
+      "num_input_tokens_seen": 860400,
+      "step": 126
+    },
+    {
+      "epoch": 3.2774193548387096,
+      "grad_norm": 0.05267687141895294,
+      "learning_rate": 1.3650237506511333e-06,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 867440,
+      "step": 127
+    },
+    {
+      "epoch": 3.303225806451613,
+      "grad_norm": 0.11772596091032028,
+      "learning_rate": 1.3263210930352737e-06,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 874256,
+      "step": 128
+    },
+    {
+      "epoch": 3.329032258064516,
+      "grad_norm": 0.9182859063148499,
+      "learning_rate": 1.2879759493841577e-06,
+      "loss": 0.008,
+      "num_input_tokens_seen": 881168,
+      "step": 129
+    },
+    {
+      "epoch": 3.3548387096774195,
+      "grad_norm": 0.07012775540351868,
+      "learning_rate": 1.2500000000000007e-06,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 888128,
+      "step": 130
+    },
+    {
+      "epoch": 3.3806451612903228,
+      "grad_norm": 0.979766309261322,
+      "learning_rate": 1.2124048127248644e-06,
+      "loss": 0.0049,
+      "num_input_tokens_seen": 895296,
+      "step": 131
+    },
+    {
+      "epoch": 3.4064516129032256,
+      "grad_norm": 0.25211745500564575,
+      "learning_rate": 1.1752018394169882e-06,
+      "loss": 0.0012,
+      "num_input_tokens_seen": 901984,
+      "step": 132
+    },
+    {
+      "epoch": 3.432258064516129,
+      "grad_norm": 0.7576802968978882,
+      "learning_rate": 1.1384024124624324e-06,
+      "loss": 0.0044,
+      "num_input_tokens_seen": 909104,
+      "step": 133
+    },
+    {
+      "epoch": 3.458064516129032,
+      "grad_norm": 0.6242833733558655,
+      "learning_rate": 1.1020177413231334e-06,
+      "loss": 0.0017,
+      "num_input_tokens_seen": 916032,
+      "step": 134
+    },
+    {
+      "epoch": 3.4838709677419355,
+      "grad_norm": 0.06409750878810883,
+      "learning_rate": 1.0660589091223854e-06,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 922736,
+      "step": 135
+    },
+    {
+      "epoch": 3.509677419354839,
+      "grad_norm": 0.7863635420799255,
+      "learning_rate": 1.0305368692688175e-06,
+      "loss": 0.0099,
+      "num_input_tokens_seen": 929472,
+      "step": 136
+    },
+    {
+      "epoch": 3.535483870967742,
+      "grad_norm": 0.7082129716873169,
+      "learning_rate": 9.95462442119879e-07,
+      "loss": 0.0068,
+      "num_input_tokens_seen": 936112,
+      "step": 137
+    },
+    {
+      "epoch": 3.5612903225806454,
+      "grad_norm": 0.5010254383087158,
+      "learning_rate": 9.608463116858544e-07,
+      "loss": 0.0025,
+      "num_input_tokens_seen": 942912,
+      "step": 138
+    },
+    {
+      "epoch": 3.587096774193548,
+      "grad_norm": 0.1011819988489151,
+      "learning_rate": 9.266990223754069e-07,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 949984,
+      "step": 139
+    },
+    {
+      "epoch": 3.6129032258064515,
+      "grad_norm": 4.317319869995117,
+      "learning_rate": 8.930309757836517e-07,
+      "loss": 0.0101,
+      "num_input_tokens_seen": 956608,
+      "step": 140
+    },
+    {
+      "epoch": 3.638709677419355,
+      "grad_norm": 0.7055479288101196,
+      "learning_rate": 8.598524275237321e-07,
+      "loss": 0.0068,
+      "num_input_tokens_seen": 963488,
+      "step": 141
+    },
+    {
+      "epoch": 3.664516129032258,
+      "grad_norm": 0.10063151270151138,
+      "learning_rate": 8.271734841028553e-07,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 970128,
+      "step": 142
+    },
+    {
+      "epoch": 3.6903225806451614,
+      "grad_norm": 3.3077032566070557,
+      "learning_rate": 7.950040998437541e-07,
+      "loss": 0.0161,
+      "num_input_tokens_seen": 976800,
+      "step": 143
+    },
+    {
+      "epoch": 3.7161290322580647,
+      "grad_norm": 1.0670413970947266,
+      "learning_rate": 7.633540738525066e-07,
+      "loss": 0.0115,
+      "num_input_tokens_seen": 984000,
+      "step": 144
+    },
+    {
+      "epoch": 3.741935483870968,
+      "grad_norm": 1.591837763786316,
+      "learning_rate": 7.322330470336314e-07,
+      "loss": 0.0052,
+      "num_input_tokens_seen": 990896,
+      "step": 145
+    },
+    {
+      "epoch": 3.767741935483871,
+      "grad_norm": 3.105447769165039,
+      "learning_rate": 7.016504991533727e-07,
+      "loss": 0.0098,
+      "num_input_tokens_seen": 997824,
+      "step": 146
+    },
+    {
+      "epoch": 3.793548387096774,
+      "grad_norm": 0.0832144096493721,
+      "learning_rate": 6.716157459520739e-07,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1004928,
+      "step": 147
+    },
+    {
+      "epoch": 3.8193548387096774,
+      "grad_norm": 0.2546854317188263,
+      "learning_rate": 6.421379363065142e-07,
+      "loss": 0.0012,
+      "num_input_tokens_seen": 1011696,
+      "step": 148
+    },
+    {
+      "epoch": 3.8451612903225807,
+      "grad_norm": 0.40274935960769653,
+      "learning_rate": 6.1322604944307e-07,
+      "loss": 0.0013,
+      "num_input_tokens_seen": 1018368,
+      "step": 149
+    },
+    {
+      "epoch": 3.870967741935484,
+      "grad_norm": 0.04935682937502861,
+      "learning_rate": 5.848888922025553e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1025424,
+      "step": 150
+    },
+    {
+      "epoch": 3.896774193548387,
+      "grad_norm": 0.8725687265396118,
+      "learning_rate": 5.571350963575728e-07,
+      "loss": 0.0026,
+      "num_input_tokens_seen": 1032016,
+      "step": 151
+    },
+    {
+      "epoch": 3.9225806451612906,
+      "grad_norm": 0.9522215127944946,
+      "learning_rate": 5.299731159831953e-07,
+      "loss": 0.0097,
+      "num_input_tokens_seen": 1038928,
+      "step": 152
+    },
+    {
+      "epoch": 3.9483870967741934,
+      "grad_norm": 2.608211040496826,
+      "learning_rate": 5.034112248817685e-07,
+      "loss": 0.0047,
+      "num_input_tokens_seen": 1045664,
+      "step": 153
+    },
+    {
+      "epoch": 3.9741935483870967,
+      "grad_norm": 0.913472592830658,
+      "learning_rate": 4.774575140626317e-07,
+      "loss": 0.0081,
+      "num_input_tokens_seen": 1052640,
+      "step": 154
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.5549633502960205,
+      "learning_rate": 4.5211988927752026e-07,
+      "loss": 0.0018,
+      "num_input_tokens_seen": 1059840,
+      "step": 155
+    },
+    {
+      "epoch": 4.025806451612903,
+      "grad_norm": 0.6482688784599304,
+      "learning_rate": 4.27406068612396e-07,
+      "loss": 0.0053,
+      "num_input_tokens_seen": 1066704,
+      "step": 156
+    },
+    {
+      "epoch": 4.051612903225807,
+      "grad_norm": 0.11119811981916428,
+      "learning_rate": 4.033235801364402e-07,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1073600,
+      "step": 157
+    },
+    {
+      "epoch": 4.077419354838709,
+      "grad_norm": 0.009617508389055729,
+      "learning_rate": 3.798797596089351e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1080208,
+      "step": 158
+    },
+    {
+      "epoch": 4.103225806451613,
+      "grad_norm": 0.18060296773910522,
+      "learning_rate": 3.5708174824471947e-07,
+      "loss": 0.0018,
+      "num_input_tokens_seen": 1086880,
+      "step": 159
+    },
+    {
+      "epoch": 4.129032258064516,
+      "grad_norm": 0.3545215427875519,
+      "learning_rate": 3.3493649053890325e-07,
+      "loss": 0.001,
+      "num_input_tokens_seen": 1093520,
+      "step": 160
+    },
+    {
+      "epoch": 4.15483870967742,
+      "grad_norm": 0.03870348632335663,
+      "learning_rate": 3.134507321515107e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1100528,
+      "step": 161
+    },
+    {
+      "epoch": 4.180645161290323,
+      "grad_norm": 0.20078836381435394,
+      "learning_rate": 2.9263101785268253e-07,
+      "loss": 0.0012,
+      "num_input_tokens_seen": 1107520,
+      "step": 162
+    },
+    {
+      "epoch": 4.2064516129032254,
+      "grad_norm": 0.00893727969378233,
+      "learning_rate": 2.7248368952908055e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1114208,
+      "step": 163
+    },
+    {
+      "epoch": 4.232258064516129,
+      "grad_norm": 0.027001328766345978,
+      "learning_rate": 2.53014884252083e-07,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1120912,
+      "step": 164
+    },
+    {
+      "epoch": 4.258064516129032,
+      "grad_norm": 0.01514577865600586,
+      "learning_rate": 2.3423053240837518e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1127872,
+      "step": 165
+    },
+    {
+      "epoch": 4.283870967741936,
+      "grad_norm": 0.0690593495965004,
+      "learning_rate": 2.1613635589349756e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1134352,
+      "step": 166
+    },
+    {
+      "epoch": 4.309677419354839,
+      "grad_norm": 0.021051278337836266,
+      "learning_rate": 1.9873786636889908e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1141440,
+      "step": 167
+    },
+    {
+      "epoch": 4.335483870967742,
+      "grad_norm": 0.3527485132217407,
+      "learning_rate": 1.8204036358303173e-07,
+      "loss": 0.0007,
+      "num_input_tokens_seen": 1148096,
+      "step": 168
+    },
+    {
+      "epoch": 4.361290322580645,
+      "grad_norm": 0.09119685739278793,
+      "learning_rate": 1.6604893375699594e-07,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1154672,
+      "step": 169
+    },
+    {
+      "epoch": 4.387096774193548,
+      "grad_norm": 0.014869263395667076,
+      "learning_rate": 1.507684480352292e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1161744,
+      "step": 170
+    },
+    {
+      "epoch": 4.412903225806452,
+      "grad_norm": 0.16110821068286896,
+      "learning_rate": 1.362035610017079e-07,
+      "loss": 0.0008,
+      "num_input_tokens_seen": 1168944,
+      "step": 171
+    },
+    {
+      "epoch": 4.438709677419355,
+      "grad_norm": 0.05959083512425423,
+      "learning_rate": 1.223587092621162e-07,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1175792,
+      "step": 172
+    },
+    {
+      "epoch": 4.464516129032258,
+      "grad_norm": 0.15646237134933472,
+      "learning_rate": 1.0923811009241142e-07,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1182800,
+      "step": 173
+    },
+    {
+      "epoch": 4.490322580645161,
+      "grad_norm": 0.024713285267353058,
+      "learning_rate": 9.684576015420277e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1189568,
+      "step": 174
+    },
+    {
+      "epoch": 4.516129032258064,
+      "grad_norm": 0.0995955690741539,
+      "learning_rate": 8.518543427732951e-08,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1196480,
+      "step": 175
+    },
+    {
+      "epoch": 4.541935483870968,
+      "grad_norm": 0.5925021767616272,
+      "learning_rate": 7.426068431000883e-08,
+      "loss": 0.0076,
+      "num_input_tokens_seen": 1203728,
+      "step": 176
+    },
+    {
+      "epoch": 4.567741935483871,
+      "grad_norm": 0.07701185345649719,
+      "learning_rate": 6.407483803691216e-08,
+      "loss": 0.0004,
+      "num_input_tokens_seen": 1210320,
+      "step": 177
+    },
+    {
+      "epoch": 4.593548387096774,
+      "grad_norm": 0.5693293213844299,
+      "learning_rate": 5.463099816548578e-08,
+      "loss": 0.004,
+      "num_input_tokens_seen": 1217248,
+      "step": 178
+    },
+    {
+      "epoch": 4.619354838709677,
+      "grad_norm": 0.025556327775120735,
+      "learning_rate": 4.593204138084006e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1224192,
+      "step": 179
+    },
+    {
+      "epoch": 4.645161290322581,
+      "grad_norm": 0.14782366156578064,
+      "learning_rate": 3.798061746947995e-08,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1230896,
+      "step": 180
+    },
+    {
+      "epoch": 4.670967741935484,
+      "grad_norm": 0.06809692084789276,
+      "learning_rate": 3.077914851215585e-08,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1237840,
+      "step": 181
+    },
+    {
+      "epoch": 4.6967741935483875,
+      "grad_norm": 0.01677132397890091,
+      "learning_rate": 2.4329828146074096e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1244688,
+      "step": 182
+    },
+    {
+      "epoch": 4.72258064516129,
+      "grad_norm": 0.019206374883651733,
+      "learning_rate": 1.8634620896695044e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1251584,
+      "step": 183
+    },
+    {
+      "epoch": 4.748387096774193,
+      "grad_norm": 0.02283065766096115,
+      "learning_rate": 1.3695261579316776e-08,
+      "loss": 0.0001,
+      "num_input_tokens_seen": 1258448,
+      "step": 184
+    },
+    {
+      "epoch": 4.774193548387097,
+      "grad_norm": 1.8589617013931274,
+      "learning_rate": 9.513254770636138e-09,
+      "loss": 0.0167,
+      "num_input_tokens_seen": 1265440,
+      "step": 185
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 0.03213285654783249,
+      "learning_rate": 6.089874350439507e-09,
+      "loss": 0.0002,
+      "num_input_tokens_seen": 1272080,
+      "step": 186
+    },
+    {
+      "epoch": 4.825806451612904,
+      "grad_norm": 0.11905555427074432,
+      "learning_rate": 3.4261631135654174e-09,
+      "loss": 0.0003,
+      "num_input_tokens_seen": 1278976,
+      "step": 187
+    },
+    {
+      "epoch": 4.851612903225806,
+      "grad_norm": 0.384997695684433,
+      "learning_rate": 1.5229324522605949e-09,
+      "loss": 0.0024,
+      "num_input_tokens_seen": 1285760,
+      "step": 188
+    },
+    {
+      "epoch": 4.877419354838709,
+      "grad_norm": 0.08503298461437225,
+      "learning_rate": 3.8076210902182607e-10,
+      "loss": 0.0005,
+      "num_input_tokens_seen": 1292368,
+      "step": 189
+    },
+    {
+      "epoch": 4.903225806451613,
+      "grad_norm": 0.4136364758014679,
+      "learning_rate": 0.0,
+      "loss": 0.0017,
+      "num_input_tokens_seen": 1299120,
+      "step": 190
+    },
+    {
+      "epoch": 4.903225806451613,
+      "num_input_tokens_seen": 1299120,
+      "step": 190,
+      "total_flos": 5.150239379108659e+16,
+      "train_loss": 0.3450760219912857,
+      "train_runtime": 2142.6975,
+      "train_samples_per_second": 11.57,
+      "train_steps_per_second": 0.089
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 190,
+  "num_input_tokens_seen": 1299120,
+  "num_train_epochs": 5,
+  "save_steps": 1000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 5.150239379108659e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fea12da3624a50d04db32839a9390a4c8041b3756454a76a3011387f179a4228
+size 6584