Push to HF hub

.gitattributes

@@ -30,3 +30,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,3 @@
+checkpoint-*/
+*/pilot_*/
+pilot_*/

all_results.json

@@ -0,0 +1,14 @@
+{
+    "epoch": 0.4,
+    "eval_loss": 1.0858668088912964,
+    "eval_runtime": 2658.3456,
+    "eval_samples": 5000,
+    "eval_samples_per_second": 9.435,
+    "eval_steps_per_second": 4.718,
+    "perplexity": 2.9620061996078197,
+    "train_loss": 1.1317986328125,
+    "train_runtime": 62833.7719,
+    "train_samples": 100000,
+    "train_samples_per_second": 3.183,
+    "train_steps_per_second": 0.398
+}

config.json

@@ -0,0 +1,39 @@
+{
+  "_name_or_path": "bigscience/bloom-560m",
+  "adapters": {
+    "adapters": {},
+    "config_map": {},
+    "fusion_config_map": {},
+    "fusions": {}
+  },
+  "apply_residual_connection_post_layernorm": false,
+  "architectures": [
+    "BloomForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "attention_softmax_in_fp32": true,
+  "bias_dropout_fusion": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_dropout": 0.0,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "masked_softmax_fusion": true,
+  "model_type": "bloom",
+  "n_head": 16,
+  "n_inner": null,
+  "n_layer": 24,
+  "offset_alibi": 100,
+  "pad_token_id": 3,
+  "pretraining_tp": 1,
+  "seq_length": 2048,
+  "skip_bias_add": true,
+  "skip_bias_add_qkv": false,
+  "slow_but_exact": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.20.0.dev0",
+  "unk_token_id": 0,
+  "use_cache": true,
+  "vocab_size": 250880
+}

eval_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 0.4,
+    "eval_loss": 1.0858668088912964,
+    "eval_runtime": 2658.3456,
+    "eval_samples": 5000,
+    "eval_samples_per_second": 9.435,
+    "eval_steps_per_second": 4.718,
+    "perplexity": 2.9620061996078197
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:831b1de926f2d80188feb2ba167e4a2534b6f28baa49afc4e3f6fc19534d34bd
+size 2236955191

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>"}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f6efc66e73f1fd69da4f436e48befb519fdff3fe18910850c1d41bd862293a5
+size 14500443

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "pad_token": "<pad>", "add_prefix_space": false, "name_or_path": "bigscience/bloom-560m", "special_tokens_map_file": null, "padding_side": "left", "tokenizer_class": "BloomTokenizer"}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.4,
+    "train_loss": 1.1317986328125,
+    "train_runtime": 62833.7719,
+    "train_samples": 100000,
+    "train_samples_per_second": 3.183,
+    "train_steps_per_second": 0.398
+}

trainer_state.json

@@ -0,0 +1,125 @@
+{
+  "best_metric": 1.0858668088912964,
+  "best_model_checkpoint": "/home/khalid/Documents/github_rep/bigscience/data/processed/50/bloom-560m_my_bitfit_100000samples_-1vocab_original-frozen/checkpoint-25000",
+  "epoch": 0.39631741854686253,
+  "global_step": 25000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.04,
+      "learning_rate": 9e-05,
+      "loss": 1.3263,
+      "step": 2500
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 8e-05,
+      "loss": 1.1812,
+      "step": 5000
+    },
+    {
+      "epoch": 0.08,
+      "eval_loss": 1.1644270420074463,
+      "eval_runtime": 2656.1309,
+      "eval_samples_per_second": 9.443,
+      "eval_steps_per_second": 4.722,
+      "step": 5000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 7e-05,
+      "loss": 1.1436,
+      "step": 7500
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 6e-05,
+      "loss": 1.1212,
+      "step": 10000
+    },
+    {
+      "epoch": 0.16,
+      "eval_loss": 1.1189237833023071,
+      "eval_runtime": 2658.6123,
+      "eval_samples_per_second": 9.434,
+      "eval_steps_per_second": 4.717,
+      "step": 10000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 5e-05,
+      "loss": 1.1092,
+      "step": 12500
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 4e-05,
+      "loss": 1.0977,
+      "step": 15000
+    },
+    {
+      "epoch": 0.24,
+      "eval_loss": 1.0991963148117065,
+      "eval_runtime": 2656.572,
+      "eval_samples_per_second": 9.441,
+      "eval_steps_per_second": 4.721,
+      "step": 15000
+    },
+    {
+      "epoch": 0.28,
+      "learning_rate": 3e-05,
+      "loss": 1.0908,
+      "step": 17500
+    },
+    {
+      "epoch": 0.32,
+      "learning_rate": 2e-05,
+      "loss": 1.0856,
+      "step": 20000
+    },
+    {
+      "epoch": 0.32,
+      "eval_loss": 1.0892250537872314,
+      "eval_runtime": 2657.3128,
+      "eval_samples_per_second": 9.438,
+      "eval_steps_per_second": 4.719,
+      "step": 20000
+    },
+    {
+      "epoch": 0.36,
+      "learning_rate": 1e-05,
+      "loss": 1.0825,
+      "step": 22500
+    },
+    {
+      "epoch": 0.4,
+      "learning_rate": 0.0,
+      "loss": 1.08,
+      "step": 25000
+    },
+    {
+      "epoch": 0.4,
+      "eval_loss": 1.0858668088912964,
+      "eval_runtime": 2656.2605,
+      "eval_samples_per_second": 9.442,
+      "eval_steps_per_second": 4.721,
+      "step": 25000
+    },
+    {
+      "epoch": 0.4,
+      "step": 25000,
+      "total_flos": 3.714827943936e+17,
+      "train_loss": 1.1317986328125,
+      "train_runtime": 62833.7719,
+      "train_samples_per_second": 3.183,
+      "train_steps_per_second": 0.398
+    }
+  ],
+  "max_steps": 25000,
+  "num_train_epochs": 1,
+  "total_flos": 3.714827943936e+17,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a3e26ae313b3da409e678f19d5dae40bc8e5b10f5ec6460abb17602db55f645c
+size 3375

wikiann-my-results.txt

@@ -0,0 +1,8 @@
+==================================================
+Results
+==================================================
+Model: /home/khalid/Documents/github_rep/bigscience/data/processed/17/bloom-560m_my_bitfit_100000samples_-1vocab_original-frozen/
+[0.08668730650154799, 0.08438818565400845, 0.13626373626373628, 0.03582089552238806, 0.05934718100890207, 0.03144654088050314, 0.020338983050847456, 0.0736196319018405, 0.05625, 0.10933940774487473]
+6.94
+3.44
+==================================================

word_embeddings.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:174ed618237771e5906be0e8d70c568de63633f3bb5e8a1e303bbdbaeaedc1ca
+size 1027605867

word_embeddings_layernorm.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10917f86841a4f322406bd72ba2e4ae8e4780aaf462c98a76eca01e0c5fbc893
+size 9703