lisa tiny llama

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/README.md

@@ -0,0 +1,58 @@
+---
+base_model: ./output_models/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1
+tags:
+- generated_from_trainer
+datasets:
+- customized
+model-index:
+- name: ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_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. -->
+
+# ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2
+
+This model is a fine-tuned version of [./output_models/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1](https://huggingface.co/./output_models/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1) on the customized 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: 1e-05
+- train_batch_size: 50
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 400
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.06
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.37.0.dev0
+- Pytorch 2.0.1+cu118
+- Datasets 2.16.1
+- Tokenizers 0.15.0

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.333669357183503,
+    "train_runtime": 83.0293,
+    "train_samples": 16323,
+    "train_samples_per_second": 196.593,
+    "train_steps_per_second": 0.494
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "./output_models/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.37.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/README.md

@@ -0,0 +1,58 @@
+---
+base_model: ./output_models/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1
+tags:
+- generated_from_trainer
+datasets:
+- customized
+model-index:
+- name: ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_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. -->
+
+# ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2
+
+This model is a fine-tuned version of [./output_models/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1](https://huggingface.co/./output_models/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1) on the customized 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: 1e-05
+- train_batch_size: 50
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 400
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.06
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.37.0.dev0
+- Pytorch 2.0.1+cu118
+- Datasets 2.16.1
+- Tokenizers 0.15.0

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.333669357183503,
+    "train_runtime": 83.0293,
+    "train_samples": 16323,
+    "train_samples_per_second": 196.593,
+    "train_steps_per_second": 0.494
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "./output_models/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.37.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "max_length": 2048,
+  "pad_token_id": 0,
+  "transformers_version": "4.37.0.dev0"
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:337e057a0dcf82adc4763566869f0f7aa13fa883ceb5f0b20999f0bbff6bfdc2
+size 2200119864

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "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": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/tokenizer.model

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

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/tokenizer_config.json

@@ -0,0 +1,41 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "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>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.333669357183503,
+    "train_runtime": 83.0293,
+    "train_samples": 16323,
+    "train_samples_per_second": 196.593,
+    "train_steps_per_second": 0.494
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/trainer_state.json

@@ -0,0 +1,276 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 50.0,
+  "global_step": 41,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.02,
+      "learning_rate": 0.0,
+      "loss": 1.3383,
+      "step": 1
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 5.045053315100077e-07,
+      "loss": 1.3478,
+      "step": 2
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 1.0090106630200154e-06,
+      "loss": 1.3229,
+      "step": 3
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 2.5000000000000015e-06,
+      "loss": 1.3415,
+      "step": 4
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 2.3902299121570315e-06,
+      "loss": 1.3333,
+      "step": 5
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 2.282162249993895e-06,
+      "loss": 1.3228,
+      "step": 6
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 2.175867509055033e-06,
+      "loss": 1.3329,
+      "step": 7
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 2.071415028359026e-06,
+      "loss": 1.3226,
+      "step": 8
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 1.9688729451668116e-06,
+      "loss": 1.3419,
+      "step": 9
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.8683081505338485e-06,
+      "loss": 1.3294,
+      "step": 10
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 1.7697862456752273e-06,
+      "loss": 1.3262,
+      "step": 11
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.6733714991721738e-06,
+      "loss": 1.3337,
+      "step": 12
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 1.5791268050478487e-06,
+      "loss": 1.3352,
+      "step": 13
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 1.4871136417398407e-06,
+      "loss": 1.3489,
+      "step": 14
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.3973920319960664e-06,
+      "loss": 1.3161,
+      "step": 15
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 1.310020503720254e-06,
+      "loss": 1.3302,
+      "step": 16
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 1.2250560517925747e-06,
+      "loss": 1.3391,
+      "step": 17
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 1.1425541008902852e-06,
+      "loss": 1.3334,
+      "step": 18
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 1.0625684693326727e-06,
+      "loss": 1.3305,
+      "step": 19
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 9.851513339738627e-07,
+      "loss": 1.3141,
+      "step": 20
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 9.103531961664108e-07,
+      "loss": 1.3426,
+      "step": 21
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 8.382228488178639e-07,
+      "loss": 1.3304,
+      "step": 22
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 7.6880734456178e-07,
+      "loss": 1.3325,
+      "step": 23
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 7.02151965063994e-07,
+      "loss": 1.3331,
+      "step": 24
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 6.383001914841264e-07,
+      "loss": 1.3259,
+      "step": 25
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 5.772936761116027e-07,
+      "loss": 1.3336,
+      "step": 26
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 5.191722151947215e-07,
+      "loss": 1.3367,
+      "step": 27
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 4.639737229804403e-07,
+      "loss": 1.3426,
+      "step": 28
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 4.1173420698186027e-07,
+      "loss": 1.319,
+      "step": 29
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 3.6248774448952695e-07,
+      "loss": 1.3408,
+      "step": 30
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 3.1626646034186137e-07,
+      "loss": 1.3202,
+      "step": 31
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 2.7310050596923323e-07,
+      "loss": 1.3442,
+      "step": 32
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 2.330180397253473e-07,
+      "loss": 1.3481,
+      "step": 33
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 1.9604520851876196e-07,
+      "loss": 1.327,
+      "step": 34
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 1.6220613075653201e-07,
+      "loss": 1.3225,
+      "step": 35
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 1.3152288061110518e-07,
+      "loss": 1.3298,
+      "step": 36
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 1.0401547362071939e-07,
+      "loss": 1.3208,
+      "step": 37
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 7.970185363271432e-08,
+      "loss": 1.3341,
+      "step": 38
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 5.859788109825737e-08,
+      "loss": 1.3582,
+      "step": 39
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.071732272613149e-08,
+      "loss": 1.3464,
+      "step": 40
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 2.6071842502326526e-08,
+      "loss": 1.3514,
+      "step": 41
+    },
+    {
+      "epoch": 1.0,
+      "step": 41,
+      "total_flos": 5.211956289155891e+16,
+      "train_loss": 1.333669357183503,
+      "train_runtime": 83.0293,
+      "train_samples_per_second": 196.593,
+      "train_steps_per_second": 0.494
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 41,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 5.211956289155891e+16,
+  "train_batch_size": 50,
+  "trial_name": null,
+  "trial_params": null
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5981ef4c8f54dc037bea3d73b498643dfbdf45df413e7354821781f756a5e51
+size 6139

ft_tinyllama_adamw_gpt4_v2_1e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "max_length": 2048,
+  "pad_token_id": 0,
+  "transformers_version": "4.37.0.dev0"
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/README.md

@@ -0,0 +1,58 @@
+---
+base_model: /hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
+tags:
+- generated_from_trainer
+datasets:
+- customized
+model-index:
+- name: ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0
+  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. -->
+
+# ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0
+
+This model is a fine-tuned version of [/hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co//hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the customized 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: 1e-05
+- train_batch_size: 50
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 400
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.01
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.37.0.dev0
+- Pytorch 2.0.1+cu118
+- Datasets 2.16.1
+- Tokenizers 0.15.0

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.5262389551333295,
+    "train_runtime": 256.4885,
+    "train_samples": 48917,
+    "train_samples_per_second": 190.718,
+    "train_steps_per_second": 0.48
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "/hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.37.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/README.md

@@ -0,0 +1,58 @@
+---
+base_model: /hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
+tags:
+- generated_from_trainer
+datasets:
+- customized
+model-index:
+- name: ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0
+  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. -->
+
+# ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0
+
+This model is a fine-tuned version of [/hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co//hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the customized 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: 1e-05
+- train_batch_size: 50
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 400
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.01
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.37.0.dev0
+- Pytorch 2.0.1+cu118
+- Datasets 2.16.1
+- Tokenizers 0.15.0

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.5262389551333295,
+    "train_runtime": 256.4885,
+    "train_samples": 48917,
+    "train_samples_per_second": 190.718,
+    "train_steps_per_second": 0.48
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/checkpoint-123/adapter_model/README.md

@@ -0,0 +1,204 @@
+---
+library_name: peft
+base_model: /hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure 
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+
+
+### Framework versions
+
+- PEFT 0.7.2.dev0

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/checkpoint-123/adapter_model/adapter_config.json

@@ -0,0 +1,32 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "up_proj",
+    "gate_proj",
+    "o_proj",
+    "q_proj",
+    "k_proj",
+    "down_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_rslora": false
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/checkpoint-123/adapter_model/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:783605c79f817de8dfe254cac0fea70a4d751deda6032a07c262e9ca98bd696b
+size 201892728

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "/hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.37.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "max_length": 2048,
+  "pad_token_id": 0,
+  "transformers_version": "4.37.0.dev0"
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82ec6f46813b9a3d765334d300ef5c27ecb7128544e9c2d219b452973b77efd3
+size 2200119864

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/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
+  }
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/tokenizer.model

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

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/tokenizer_config.json

@@ -0,0 +1,41 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "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>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.5262389551333295,
+    "train_runtime": 256.4885,
+    "train_samples": 48917,
+    "train_samples_per_second": 190.718,
+    "train_steps_per_second": 0.48
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/trainer_state.json

@@ -0,0 +1,768 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 50.0,
+  "global_step": 123,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.0,
+      "loss": 1.6136,
+      "step": 1
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.996630220997058e-06,
+      "loss": 1.5885,
+      "step": 2
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-05,
+      "loss": 1.6267,
+      "step": 3
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.998369180404283e-06,
+      "loss": 1.605,
+      "step": 4
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.993477785446151e-06,
+      "loss": 1.6058,
+      "step": 5
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.985329005918702e-06,
+      "loss": 1.6161,
+      "step": 6
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.973928157497675e-06,
+      "loss": 1.6081,
+      "step": 7
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.959282677273869e-06,
+      "loss": 1.6148,
+      "step": 8
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.941402118901743e-06,
+      "loss": 1.5983,
+      "step": 9
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.920298146367287e-06,
+      "loss": 1.5948,
+      "step": 10
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.895984526379282e-06,
+      "loss": 1.5968,
+      "step": 11
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.868477119388897e-06,
+      "loss": 1.6115,
+      "step": 12
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.837793869243468e-06,
+      "loss": 1.6084,
+      "step": 13
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.803954791481239e-06,
+      "loss": 1.5956,
+      "step": 14
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.766981960274653e-06,
+      "loss": 1.6045,
+      "step": 15
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.726899494030768e-06,
+      "loss": 1.6136,
+      "step": 16
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.68373353965814e-06,
+      "loss": 1.5897,
+      "step": 17
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.637512255510475e-06,
+      "loss": 1.6069,
+      "step": 18
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.588265793018141e-06,
+      "loss": 1.597,
+      "step": 19
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.536026277019562e-06,
+      "loss": 1.6212,
+      "step": 20
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.480827784805278e-06,
+      "loss": 1.5897,
+      "step": 21
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.422706323888398e-06,
+      "loss": 1.6066,
+      "step": 22
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.361699808515877e-06,
+      "loss": 1.6051,
+      "step": 23
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.297848034936007e-06,
+      "loss": 1.5903,
+      "step": 24
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.231192655438222e-06,
+      "loss": 1.5953,
+      "step": 25
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.161777151182137e-06,
+      "loss": 1.5954,
+      "step": 26
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.089646803833589e-06,
+      "loss": 1.5644,
+      "step": 27
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.014848666026138e-06,
+      "loss": 1.5686,
+      "step": 28
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.937431530667329e-06,
+      "loss": 1.5862,
+      "step": 29
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.857445899109716e-06,
+      "loss": 1.5795,
+      "step": 30
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.774943948207427e-06,
+      "loss": 1.5631,
+      "step": 31
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.689979496279747e-06,
+      "loss": 1.5811,
+      "step": 32
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.602607968003935e-06,
+      "loss": 1.5675,
+      "step": 33
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.51288635826016e-06,
+      "loss": 1.5548,
+      "step": 34
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.420873194952153e-06,
+      "loss": 1.5828,
+      "step": 35
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.326628500827826e-06,
+      "loss": 1.568,
+      "step": 36
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.230213754324771e-06,
+      "loss": 1.5588,
+      "step": 37
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.131691849466154e-06,
+      "loss": 1.5432,
+      "step": 38
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 8.031127054833192e-06,
+      "loss": 1.5518,
+      "step": 39
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.928584971640974e-06,
+      "loss": 1.5502,
+      "step": 40
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.82413249094497e-06,
+      "loss": 1.5597,
+      "step": 41
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.717837750006108e-06,
+      "loss": 1.5353,
+      "step": 42
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.609770087842968e-06,
+      "loss": 1.5659,
+      "step": 43
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 1.5484,
+      "step": 44
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.388599092561315e-06,
+      "loss": 1.5166,
+      "step": 45
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.2756400354400445e-06,
+      "loss": 1.5233,
+      "step": 46
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.161196514973736e-06,
+      "loss": 1.5254,
+      "step": 47
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 7.045343185856701e-06,
+      "loss": 1.53,
+      "step": 48
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.92815562244068e-06,
+      "loss": 1.5093,
+      "step": 49
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.80971026943559e-06,
+      "loss": 1.5434,
+      "step": 50
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.690084392042514e-06,
+      "loss": 1.5221,
+      "step": 51
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.569356025551454e-06,
+      "loss": 1.5259,
+      "step": 52
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.447603924436744e-06,
+      "loss": 1.5364,
+      "step": 53
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.324907510983309e-06,
+      "loss": 1.5314,
+      "step": 54
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.2013468234773034e-06,
+      "loss": 1.5157,
+      "step": 55
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 6.077002463994908e-06,
+      "loss": 1.5075,
+      "step": 56
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.951955545823342e-06,
+      "loss": 1.5363,
+      "step": 57
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.826287640548425e-06,
+      "loss": 1.5119,
+      "step": 58
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.7000807248431466e-06,
+      "loss": 1.5325,
+      "step": 59
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.573417126992004e-06,
+      "loss": 1.5059,
+      "step": 60
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.446379473185972e-06,
+      "loss": 1.491,
+      "step": 61
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.319050633623141e-06,
+      "loss": 1.5127,
+      "step": 62
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.191513668450178e-06,
+      "loss": 1.505,
+      "step": 63
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.06385177357987e-06,
+      "loss": 1.5233,
+      "step": 64
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.936148226420131e-06,
+      "loss": 1.5071,
+      "step": 65
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.808486331549824e-06,
+      "loss": 1.514,
+      "step": 66
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.6809493663768594e-06,
+      "loss": 1.4965,
+      "step": 67
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.553620526814029e-06,
+      "loss": 1.5118,
+      "step": 68
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.426582873007999e-06,
+      "loss": 1.5003,
+      "step": 69
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.299919275156857e-06,
+      "loss": 1.5108,
+      "step": 70
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.173712359451575e-06,
+      "loss": 1.4794,
+      "step": 71
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.048044454176659e-06,
+      "loss": 1.4992,
+      "step": 72
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.922997536005094e-06,
+      "loss": 1.5076,
+      "step": 73
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.7986531765226965e-06,
+      "loss": 1.5085,
+      "step": 74
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.675092489016693e-06,
+      "loss": 1.489,
+      "step": 75
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.5523960755632573e-06,
+      "loss": 1.4943,
+      "step": 76
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.4306439744485466e-06,
+      "loss": 1.491,
+      "step": 77
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.309915607957487e-06,
+      "loss": 1.4984,
+      "step": 78
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.190289730564411e-06,
+      "loss": 1.4842,
+      "step": 79
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.071844377559322e-06,
+      "loss": 1.4969,
+      "step": 80
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.954656814143302e-06,
+      "loss": 1.475,
+      "step": 81
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.838803485026265e-06,
+      "loss": 1.4809,
+      "step": 82
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.7243599645599563e-06,
+      "loss": 1.4781,
+      "step": 83
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.611400907438687e-06,
+      "loss": 1.4609,
+      "step": 84
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.5000000000000015e-06,
+      "loss": 1.4863,
+      "step": 85
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.3902299121570315e-06,
+      "loss": 1.4806,
+      "step": 86
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.282162249993895e-06,
+      "loss": 1.4999,
+      "step": 87
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 2.175867509055033e-06,
+      "loss": 1.4633,
+      "step": 88
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 2.071415028359026e-06,
+      "loss": 1.4959,
+      "step": 89
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9688729451668116e-06,
+      "loss": 1.4753,
+      "step": 90
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.8683081505338485e-06,
+      "loss": 1.4996,
+      "step": 91
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7697862456752273e-06,
+      "loss": 1.4857,
+      "step": 92
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.6733714991721738e-06,
+      "loss": 1.4848,
+      "step": 93
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.5791268050478487e-06,
+      "loss": 1.4671,
+      "step": 94
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4871136417398407e-06,
+      "loss": 1.4601,
+      "step": 95
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.3973920319960664e-06,
+      "loss": 1.4626,
+      "step": 96
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.310020503720254e-06,
+      "loss": 1.4829,
+      "step": 97
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.2250560517925747e-06,
+      "loss": 1.4794,
+      "step": 98
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.1425541008902852e-06,
+      "loss": 1.4734,
+      "step": 99
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.0625684693326727e-06,
+      "loss": 1.4855,
+      "step": 100
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.851513339738627e-07,
+      "loss": 1.4677,
+      "step": 101
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.103531961664108e-07,
+      "loss": 1.4792,
+      "step": 102
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 8.382228488178639e-07,
+      "loss": 1.4661,
+      "step": 103
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 7.6880734456178e-07,
+      "loss": 1.4749,
+      "step": 104
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 7.02151965063994e-07,
+      "loss": 1.459,
+      "step": 105
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 6.383001914841264e-07,
+      "loss": 1.4919,
+      "step": 106
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 5.772936761116027e-07,
+      "loss": 1.4667,
+      "step": 107
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 5.191722151947215e-07,
+      "loss": 1.4719,
+      "step": 108
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.639737229804403e-07,
+      "loss": 1.4805,
+      "step": 109
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.1173420698186027e-07,
+      "loss": 1.4771,
+      "step": 110
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.6248774448952695e-07,
+      "loss": 1.4877,
+      "step": 111
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 3.1626646034186137e-07,
+      "loss": 1.4829,
+      "step": 112
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.7310050596923323e-07,
+      "loss": 1.4706,
+      "step": 113
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.330180397253473e-07,
+      "loss": 1.4575,
+      "step": 114
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.9604520851876196e-07,
+      "loss": 1.4792,
+      "step": 115
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.6220613075653201e-07,
+      "loss": 1.4818,
+      "step": 116
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.3152288061110518e-07,
+      "loss": 1.4479,
+      "step": 117
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.0401547362071939e-07,
+      "loss": 1.4813,
+      "step": 118
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 7.970185363271432e-08,
+      "loss": 1.4972,
+      "step": 119
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 5.859788109825737e-08,
+      "loss": 1.4805,
+      "step": 120
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.071732272613149e-08,
+      "loss": 1.4657,
+      "step": 121
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 2.6071842502326526e-08,
+      "loss": 1.4806,
+      "step": 122
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4670994081297796e-08,
+      "loss": 1.4792,
+      "step": 123
+    },
+    {
+      "epoch": 1.0,
+      "step": 123,
+      "total_flos": 1.7161280189733274e+17,
+      "train_loss": 1.5262389551333295,
+      "train_runtime": 256.4885,
+      "train_samples_per_second": 190.718,
+      "train_steps_per_second": 0.48
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 123,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 1.7161280189733274e+17,
+  "train_batch_size": 50,
+  "trial_name": null,
+  "trial_params": null
+}

ft_tinyllama_adamw_gpt4_v2_1e-5_lora_rank128_0/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "max_length": 2048,
+  "pad_token_id": 0,
+  "transformers_version": "4.37.0.dev0"
+}

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/README.md

@@ -0,0 +1,58 @@
+---
+base_model: /hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T
+tags:
+- generated_from_trainer
+datasets:
+- customized
+model-index:
+- name: ft_tinyllama_adamw_gpt4_v2_1e-6_0
+  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. -->
+
+# ft_tinyllama_adamw_gpt4_v2_1e-6_0
+
+This model is a fine-tuned version of [/hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T](https://huggingface.co//hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T) on the customized 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: 1e-06
+- train_batch_size: 50
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 400
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.01
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.37.0.dev0
+- Pytorch 2.0.1+cu118
+- Datasets 2.16.1
+- Tokenizers 0.15.0

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.4142228171108215,
+    "train_runtime": 520.881,
+    "train_samples": 48917,
+    "train_samples_per_second": 93.912,
+    "train_steps_per_second": 0.236
+}

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "/hpc2hdd/home/xliu886/workspace/share_models/TinyLlama/TinyLlama-1.1B-intermediate-step-1431k-3T",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.37.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "max_length": 2048,
+  "pad_token_id": 0,
+  "transformers_version": "4.37.0.dev0"
+}

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4b8456ee3da423a83830b4144692b23d1db58b25ed84f4e1ca72a496babea10
+size 2200119864

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/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
+  }
+}

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/tokenizer.model

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

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/tokenizer_config.json

@@ -0,0 +1,41 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "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>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "padding_side": "right",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.4142228171108215,
+    "train_runtime": 520.881,
+    "train_samples": 48917,
+    "train_samples_per_second": 93.912,
+    "train_steps_per_second": 0.236
+}

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/trainer_state.json

@@ -0,0 +1,768 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 50.0,
+  "global_step": 123,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.01,
+      "learning_rate": 0.0,
+      "loss": 1.6136,
+      "step": 1
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 4.996630220997057e-07,
+      "loss": 1.5885,
+      "step": 2
+    },
+    {
+      "epoch": 0.02,
+      "learning_rate": 1e-06,
+      "loss": 1.6267,
+      "step": 3
+    },
+    {
+      "epoch": 0.03,
+      "learning_rate": 9.99836918040428e-07,
+      "loss": 1.605,
+      "step": 4
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 9.993477785446149e-07,
+      "loss": 1.6055,
+      "step": 5
+    },
+    {
+      "epoch": 0.05,
+      "learning_rate": 9.985329005918702e-07,
+      "loss": 1.6141,
+      "step": 6
+    },
+    {
+      "epoch": 0.06,
+      "learning_rate": 9.973928157497674e-07,
+      "loss": 1.6058,
+      "step": 7
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.959282677273868e-07,
+      "loss": 1.6068,
+      "step": 8
+    },
+    {
+      "epoch": 0.07,
+      "learning_rate": 9.941402118901742e-07,
+      "loss": 1.5894,
+      "step": 9
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 9.920298146367286e-07,
+      "loss": 1.585,
+      "step": 10
+    },
+    {
+      "epoch": 0.09,
+      "learning_rate": 9.89598452637928e-07,
+      "loss": 1.5869,
+      "step": 11
+    },
+    {
+      "epoch": 0.1,
+      "learning_rate": 9.868477119388894e-07,
+      "loss": 1.5834,
+      "step": 12
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.837793869243467e-07,
+      "loss": 1.5773,
+      "step": 13
+    },
+    {
+      "epoch": 0.11,
+      "learning_rate": 9.803954791481238e-07,
+      "loss": 1.562,
+      "step": 14
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 9.766981960274652e-07,
+      "loss": 1.5705,
+      "step": 15
+    },
+    {
+      "epoch": 0.13,
+      "learning_rate": 9.726899494030766e-07,
+      "loss": 1.5771,
+      "step": 16
+    },
+    {
+      "epoch": 0.14,
+      "learning_rate": 9.683733539658138e-07,
+      "loss": 1.5526,
+      "step": 17
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.637512255510474e-07,
+      "loss": 1.5695,
+      "step": 18
+    },
+    {
+      "epoch": 0.15,
+      "learning_rate": 9.58826579301814e-07,
+      "loss": 1.5537,
+      "step": 19
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 9.53602627701956e-07,
+      "loss": 1.546,
+      "step": 20
+    },
+    {
+      "epoch": 0.17,
+      "learning_rate": 9.480827784805278e-07,
+      "loss": 1.5086,
+      "step": 21
+    },
+    {
+      "epoch": 0.18,
+      "learning_rate": 9.422706323888396e-07,
+      "loss": 1.5218,
+      "step": 22
+    },
+    {
+      "epoch": 0.19,
+      "learning_rate": 9.361699808515875e-07,
+      "loss": 1.5191,
+      "step": 23
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.297848034936005e-07,
+      "loss": 1.5038,
+      "step": 24
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 9.23119265543822e-07,
+      "loss": 1.5081,
+      "step": 25
+    },
+    {
+      "epoch": 0.21,
+      "learning_rate": 9.161777151182135e-07,
+      "loss": 1.5069,
+      "step": 26
+    },
+    {
+      "epoch": 0.22,
+      "learning_rate": 9.089646803833588e-07,
+      "loss": 1.4745,
+      "step": 27
+    },
+    {
+      "epoch": 0.23,
+      "learning_rate": 9.014848666026138e-07,
+      "loss": 1.4748,
+      "step": 28
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.937431530667327e-07,
+      "loss": 1.4912,
+      "step": 29
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 8.857445899109715e-07,
+      "loss": 1.4848,
+      "step": 30
+    },
+    {
+      "epoch": 0.25,
+      "learning_rate": 8.774943948207425e-07,
+      "loss": 1.4675,
+      "step": 31
+    },
+    {
+      "epoch": 0.26,
+      "learning_rate": 8.689979496279746e-07,
+      "loss": 1.4832,
+      "step": 32
+    },
+    {
+      "epoch": 0.27,
+      "learning_rate": 8.602607968003934e-07,
+      "loss": 1.4717,
+      "step": 33
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.512886358260159e-07,
+      "loss": 1.4497,
+      "step": 34
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 8.420873194952152e-07,
+      "loss": 1.4667,
+      "step": 35
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 8.326628500827825e-07,
+      "loss": 1.4455,
+      "step": 36
+    },
+    {
+      "epoch": 0.3,
+      "learning_rate": 8.230213754324771e-07,
+      "loss": 1.431,
+      "step": 37
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 8.131691849466152e-07,
+      "loss": 1.4147,
+      "step": 38
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 8.03112705483319e-07,
+      "loss": 1.4158,
+      "step": 39
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.928584971640974e-07,
+      "loss": 1.4145,
+      "step": 40
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 7.824132490944968e-07,
+      "loss": 1.4208,
+      "step": 41
+    },
+    {
+      "epoch": 0.34,
+      "learning_rate": 7.717837750006107e-07,
+      "loss": 1.397,
+      "step": 42
+    },
+    {
+      "epoch": 0.35,
+      "learning_rate": 7.609770087842967e-07,
+      "loss": 1.4244,
+      "step": 43
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 7.5e-07,
+      "loss": 1.4063,
+      "step": 44
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.388599092561314e-07,
+      "loss": 1.378,
+      "step": 45
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 7.275640035440044e-07,
+      "loss": 1.3817,
+      "step": 46
+    },
+    {
+      "epoch": 0.38,
+      "learning_rate": 7.161196514973736e-07,
+      "loss": 1.3795,
+      "step": 47
+    },
+    {
+      "epoch": 0.39,
+      "learning_rate": 7.0453431858567e-07,
+      "loss": 1.3858,
+      "step": 48
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 6.928155622440679e-07,
+      "loss": 1.3699,
+      "step": 49
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.809710269435589e-07,
+      "loss": 1.3947,
+      "step": 50
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 6.690084392042513e-07,
+      "loss": 1.3767,
+      "step": 51
+    },
+    {
+      "epoch": 0.42,
+      "learning_rate": 6.569356025551454e-07,
+      "loss": 1.3831,
+      "step": 52
+    },
+    {
+      "epoch": 0.43,
+      "learning_rate": 6.447603924436743e-07,
+      "loss": 1.3905,
+      "step": 53
+    },
+    {
+      "epoch": 0.44,
+      "learning_rate": 6.324907510983309e-07,
+      "loss": 1.3874,
+      "step": 54
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 6.201346823477302e-07,
+      "loss": 1.3749,
+      "step": 55
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 6.077002463994907e-07,
+      "loss": 1.3649,
+      "step": 56
+    },
+    {
+      "epoch": 0.46,
+      "learning_rate": 5.951955545823342e-07,
+      "loss": 1.392,
+      "step": 57
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 5.826287640548424e-07,
+      "loss": 1.3692,
+      "step": 58
+    },
+    {
+      "epoch": 0.48,
+      "learning_rate": 5.700080724843146e-07,
+      "loss": 1.3866,
+      "step": 59
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 5.573417126992002e-07,
+      "loss": 1.363,
+      "step": 60
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.446379473185971e-07,
+      "loss": 1.3512,
+      "step": 61
+    },
+    {
+      "epoch": 0.5,
+      "learning_rate": 5.319050633623141e-07,
+      "loss": 1.3702,
+      "step": 62
+    },
+    {
+      "epoch": 0.51,
+      "learning_rate": 5.191513668450177e-07,
+      "loss": 1.3669,
+      "step": 63
+    },
+    {
+      "epoch": 0.52,
+      "learning_rate": 5.063851773579869e-07,
+      "loss": 1.3829,
+      "step": 64
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 4.936148226420131e-07,
+      "loss": 1.3668,
+      "step": 65
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.808486331549823e-07,
+      "loss": 1.3743,
+      "step": 66
+    },
+    {
+      "epoch": 0.54,
+      "learning_rate": 4.6809493663768586e-07,
+      "loss": 1.3565,
+      "step": 67
+    },
+    {
+      "epoch": 0.55,
+      "learning_rate": 4.5536205268140286e-07,
+      "loss": 1.3749,
+      "step": 68
+    },
+    {
+      "epoch": 0.56,
+      "learning_rate": 4.4265828730079977e-07,
+      "loss": 1.3612,
+      "step": 69
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 4.2999192751568557e-07,
+      "loss": 1.3707,
+      "step": 70
+    },
+    {
+      "epoch": 0.58,
+      "learning_rate": 4.1737123594515744e-07,
+      "loss": 1.3423,
+      "step": 71
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 4.0480444541766586e-07,
+      "loss": 1.365,
+      "step": 72
+    },
+    {
+      "epoch": 0.59,
+      "learning_rate": 3.9229975360050934e-07,
+      "loss": 1.3703,
+      "step": 73
+    },
+    {
+      "epoch": 0.6,
+      "learning_rate": 3.798653176522696e-07,
+      "loss": 1.3686,
+      "step": 74
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 3.6750924890166923e-07,
+      "loss": 1.3512,
+      "step": 75
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.552396075563257e-07,
+      "loss": 1.3549,
+      "step": 76
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.4306439744485463e-07,
+      "loss": 1.3539,
+      "step": 77
+    },
+    {
+      "epoch": 0.63,
+      "learning_rate": 3.3099156079574867e-07,
+      "loss": 1.3624,
+      "step": 78
+    },
+    {
+      "epoch": 0.64,
+      "learning_rate": 3.1902897305644103e-07,
+      "loss": 1.3502,
+      "step": 79
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 3.0718443775593215e-07,
+      "loss": 1.3599,
+      "step": 80
+    },
+    {
+      "epoch": 0.66,
+      "learning_rate": 2.954656814143301e-07,
+      "loss": 1.3371,
+      "step": 81
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.8388034850262646e-07,
+      "loss": 1.3438,
+      "step": 82
+    },
+    {
+      "epoch": 0.67,
+      "learning_rate": 2.724359964559956e-07,
+      "loss": 1.3454,
+      "step": 83
+    },
+    {
+      "epoch": 0.68,
+      "learning_rate": 2.6114009074386865e-07,
+      "loss": 1.3262,
+      "step": 84
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 2.500000000000001e-07,
+      "loss": 1.3476,
+      "step": 85
+    },
+    {
+      "epoch": 0.7,
+      "learning_rate": 2.390229912157031e-07,
+      "loss": 1.3433,
+      "step": 86
+    },
+    {
+      "epoch": 0.71,
+      "learning_rate": 2.2821622499938948e-07,
+      "loss": 1.3638,
+      "step": 87
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 2.1758675090550328e-07,
+      "loss": 1.3294,
+      "step": 88
+    },
+    {
+      "epoch": 0.72,
+      "learning_rate": 2.071415028359026e-07,
+      "loss": 1.3606,
+      "step": 89
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 1.9688729451668111e-07,
+      "loss": 1.3422,
+      "step": 90
+    },
+    {
+      "epoch": 0.74,
+      "learning_rate": 1.8683081505338484e-07,
+      "loss": 1.3651,
+      "step": 91
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.7697862456752271e-07,
+      "loss": 1.3507,
+      "step": 92
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.6733714991721738e-07,
+      "loss": 1.3511,
+      "step": 93
+    },
+    {
+      "epoch": 0.76,
+      "learning_rate": 1.5791268050478483e-07,
+      "loss": 1.3327,
+      "step": 94
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 1.4871136417398405e-07,
+      "loss": 1.3329,
+      "step": 95
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 1.3973920319960665e-07,
+      "loss": 1.3309,
+      "step": 96
+    },
+    {
+      "epoch": 0.79,
+      "learning_rate": 1.310020503720254e-07,
+      "loss": 1.3495,
+      "step": 97
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.2250560517925745e-07,
+      "loss": 1.3511,
+      "step": 98
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.142554100890285e-07,
+      "loss": 1.3382,
+      "step": 99
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 1.0625684693326725e-07,
+      "loss": 1.3504,
+      "step": 100
+    },
+    {
+      "epoch": 0.82,
+      "learning_rate": 9.851513339738626e-08,
+      "loss": 1.3336,
+      "step": 101
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 9.103531961664107e-08,
+      "loss": 1.3454,
+      "step": 102
+    },
+    {
+      "epoch": 0.84,
+      "learning_rate": 8.382228488178638e-08,
+      "loss": 1.3335,
+      "step": 103
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 7.688073445617798e-08,
+      "loss": 1.345,
+      "step": 104
+    },
+    {
+      "epoch": 0.85,
+      "learning_rate": 7.02151965063994e-08,
+      "loss": 1.3302,
+      "step": 105
+    },
+    {
+      "epoch": 0.86,
+      "learning_rate": 6.383001914841263e-08,
+      "loss": 1.3573,
+      "step": 106
+    },
+    {
+      "epoch": 0.87,
+      "learning_rate": 5.772936761116026e-08,
+      "loss": 1.3354,
+      "step": 107
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 5.191722151947214e-08,
+      "loss": 1.3423,
+      "step": 108
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.639737229804402e-08,
+      "loss": 1.3474,
+      "step": 109
+    },
+    {
+      "epoch": 0.89,
+      "learning_rate": 4.117342069818602e-08,
+      "loss": 1.3471,
+      "step": 110
+    },
+    {
+      "epoch": 0.9,
+      "learning_rate": 3.624877444895269e-08,
+      "loss": 1.3548,
+      "step": 111
+    },
+    {
+      "epoch": 0.91,
+      "learning_rate": 3.162664603418613e-08,
+      "loss": 1.3511,
+      "step": 112
+    },
+    {
+      "epoch": 0.92,
+      "learning_rate": 2.731005059692332e-08,
+      "loss": 1.34,
+      "step": 113
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 2.3301803972534728e-08,
+      "loss": 1.3252,
+      "step": 114
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 1.9604520851876194e-08,
+      "loss": 1.3459,
+      "step": 115
+    },
+    {
+      "epoch": 0.94,
+      "learning_rate": 1.62206130756532e-08,
+      "loss": 1.3523,
+      "step": 116
+    },
+    {
+      "epoch": 0.95,
+      "learning_rate": 1.3152288061110517e-08,
+      "loss": 1.317,
+      "step": 117
+    },
+    {
+      "epoch": 0.96,
+      "learning_rate": 1.0401547362071938e-08,
+      "loss": 1.3519,
+      "step": 118
+    },
+    {
+      "epoch": 0.97,
+      "learning_rate": 7.970185363271431e-09,
+      "loss": 1.3668,
+      "step": 119
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 5.859788109825736e-09,
+      "loss": 1.3482,
+      "step": 120
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.071732272613148e-09,
+      "loss": 1.3382,
+      "step": 121
+    },
+    {
+      "epoch": 0.99,
+      "learning_rate": 2.6071842502326526e-09,
+      "loss": 1.3467,
+      "step": 122
+    },
+    {
+      "epoch": 1.0,
+      "learning_rate": 1.4670994081297795e-09,
+      "loss": 1.348,
+      "step": 123
+    },
+    {
+      "epoch": 1.0,
+      "step": 123,
+      "total_flos": 1.5635868867467674e+17,
+      "train_loss": 1.4142228171108215,
+      "train_runtime": 520.881,
+      "train_samples_per_second": 93.912,
+      "train_steps_per_second": 0.236
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 123,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 1.5635868867467674e+17,
+  "train_batch_size": 50,
+  "trial_name": null,
+  "trial_params": null
+}

ft_tinyllama_adamw_gpt4_v2_1e-6_0/ft_tinyllama_adamw_gpt4_v2_1e-6_0/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f76ed811a4a4dabe30f0f2eb6b84ca537e9e42346d60324d1884ec4fe562839b
+size 6011

ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/README.md

@@ -0,0 +1,58 @@
+---
+base_model: ./output_models/ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1
+tags:
+- generated_from_trainer
+datasets:
+- customized
+model-index:
+- name: ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_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. -->
+
+# ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2
+
+This model is a fine-tuned version of [./output_models/ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1](https://huggingface.co/./output_models/ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1) on the customized 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-05
+- train_batch_size: 50
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 400
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.06
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.37.0.dev0
+- Pytorch 2.0.1+cu118
+- Datasets 2.16.1
+- Tokenizers 0.15.0

ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 1.0,
+    "train_loss": 1.226966727070692,
+    "train_runtime": 86.6177,
+    "train_samples": 16323,
+    "train_samples_per_second": 188.449,
+    "train_steps_per_second": 0.473
+}

ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "./output_models/ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 5632,
+  "max_position_embeddings": 2048,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 22,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.37.0.dev0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2/README.md

@@ -0,0 +1,58 @@
+---
+base_model: ./output_models/ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1
+tags:
+- generated_from_trainer
+datasets:
+- customized
+model-index:
+- name: ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_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. -->
+
+# ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_2
+
+This model is a fine-tuned version of [./output_models/ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1](https://huggingface.co/./output_models/ft_tinyllama_adamw_gpt4_v2_5e-5_activate_random*2_no_1_k3_warm0.06_openheadAndemb_1) on the customized 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-05
+- train_batch_size: 50
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 8
+- total_train_batch_size: 400
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_ratio: 0.06
+- num_epochs: 1.0
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.37.0.dev0
+- Pytorch 2.0.1+cu118
+- Datasets 2.16.1
+- Tokenizers 0.15.0