Training in progress, step 6500, checkpoint

checkpoint-6500/config.json

@@ -0,0 +1,33 @@
+{
+  "activation_function": "gelu_new",
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 1,
+  "dtype": "float32",
+  "embd_pdrop": 0.1,
+  "eos_token_id": 2,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "pad_token_id": 2,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "transformers_version": "4.56.1",
+  "use_cache": true,
+  "vocab_size": 30000
+}

checkpoint-6500/generation_config.json

@@ -0,0 +1,9 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": [
+    2
+  ],
+  "pad_token_id": 2,
+  "transformers_version": "4.56.1"
+}

checkpoint-6500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc45b357c9b45d2e493fb302ea64a097510996b4bd37340c5e1a2651fedf2c98
+size 435544704

checkpoint-6500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55616da341ecfa2f938ba385ce94625346e9df3acc5326be19934a3bc2b8e117
+size 871183627

checkpoint-6500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48f97a02e0ee100fc23fe9449110ef7de5e799597d4d363230684342e730c0df
+size 14709

checkpoint-6500/scaler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:72ec040f7dbfafb8a0d9c346b901cc53a9760e25574232230ec90bf3c4a5f84d
+size 1383

checkpoint-6500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a822a85c118b31ab6428db5e2a58ab500774c300151893ec932e938f067b72c7
+size 1465

checkpoint-6500/special_tokens_map.json

@@ -0,0 +1,8 @@
+{
+  "bos_token": "<s>",
+  "eos_token": "</s>",
+  "mask_token": "<mask>",
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "unk_token": "<unk>"
+}

checkpoint-6500/tokenizer_config.json

@@ -0,0 +1,62 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<pad>",
+      "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
+    },
+    "3": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "4": {
+      "content": "<mask>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "5": {
+      "content": "<sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<s>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "extra_special_tokens": {},
+  "mask_token": "<mask>",
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "</s>",
+  "sep_token": "<sep>",
+  "tokenizer_class": "PreTrainedTokenizerFast",
+  "unk_token": "<unk>"
+}

checkpoint-6500/trainer_state.json

@@ -0,0 +1,266 @@
+{
+  "best_global_step": 4220,
+  "best_metric": 4.378113269805908,
+  "best_model_checkpoint": null,
+  "epoch": 15.402843601895734,
+  "eval_steps": 500,
+  "global_step": 6500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.002369668246445498,
+      "grad_norm": 20.34552001953125,
+      "learning_rate": 0.0,
+      "loss": 10.3309,
+      "step": 1
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 1.0193040370941162,
+      "learning_rate": 8.42e-05,
+      "loss": 5.762,
+      "step": 422
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 4.781050205230713,
+      "eval_runtime": 21.3156,
+      "eval_samples_per_second": 1292.106,
+      "eval_steps_per_second": 5.067,
+      "step": 422
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6101345419883728,
+      "learning_rate": 9.566919191919192e-05,
+      "loss": 4.6298,
+      "step": 844
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 4.602915287017822,
+      "eval_runtime": 22.4738,
+      "eval_samples_per_second": 1225.514,
+      "eval_steps_per_second": 4.806,
+      "step": 844
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 0.5042136907577515,
+      "learning_rate": 9.03409090909091e-05,
+      "loss": 4.5095,
+      "step": 1266
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 4.522061347961426,
+      "eval_runtime": 31.9704,
+      "eval_samples_per_second": 861.483,
+      "eval_steps_per_second": 3.378,
+      "step": 1266
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 0.5082475543022156,
+      "learning_rate": 8.501262626262628e-05,
+      "loss": 4.4388,
+      "step": 1688
+    },
+    {
+      "epoch": 4.0,
+      "eval_loss": 4.47515869140625,
+      "eval_runtime": 31.848,
+      "eval_samples_per_second": 864.796,
+      "eval_steps_per_second": 3.391,
+      "step": 1688
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.5105661749839783,
+      "learning_rate": 7.968434343434343e-05,
+      "loss": 4.3851,
+      "step": 2110
+    },
+    {
+      "epoch": 5.0,
+      "eval_loss": 4.442058563232422,
+      "eval_runtime": 32.0508,
+      "eval_samples_per_second": 859.323,
+      "eval_steps_per_second": 3.37,
+      "step": 2110
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 0.5460625886917114,
+      "learning_rate": 7.435606060606061e-05,
+      "loss": 4.3391,
+      "step": 2532
+    },
+    {
+      "epoch": 6.0,
+      "eval_loss": 4.4180145263671875,
+      "eval_runtime": 30.8915,
+      "eval_samples_per_second": 891.572,
+      "eval_steps_per_second": 3.496,
+      "step": 2532
+    },
+    {
+      "epoch": 7.0,
+      "grad_norm": 0.5497872233390808,
+      "learning_rate": 6.902777777777779e-05,
+      "loss": 4.2975,
+      "step": 2954
+    },
+    {
+      "epoch": 7.0,
+      "eval_loss": 4.39995813369751,
+      "eval_runtime": 32.8624,
+      "eval_samples_per_second": 838.1,
+      "eval_steps_per_second": 3.286,
+      "step": 2954
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.5838669538497925,
+      "learning_rate": 6.369949494949495e-05,
+      "loss": 4.258,
+      "step": 3376
+    },
+    {
+      "epoch": 8.0,
+      "eval_loss": 4.387345790863037,
+      "eval_runtime": 30.4389,
+      "eval_samples_per_second": 904.829,
+      "eval_steps_per_second": 3.548,
+      "step": 3376
+    },
+    {
+      "epoch": 9.0,
+      "grad_norm": 0.6011605262756348,
+      "learning_rate": 5.837121212121213e-05,
+      "loss": 4.2193,
+      "step": 3798
+    },
+    {
+      "epoch": 9.0,
+      "eval_loss": 4.380561351776123,
+      "eval_runtime": 31.2627,
+      "eval_samples_per_second": 880.986,
+      "eval_steps_per_second": 3.455,
+      "step": 3798
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 0.6363890171051025,
+      "learning_rate": 5.30429292929293e-05,
+      "loss": 4.1807,
+      "step": 4220
+    },
+    {
+      "epoch": 10.0,
+      "eval_loss": 4.378113269805908,
+      "eval_runtime": 31.7839,
+      "eval_samples_per_second": 866.54,
+      "eval_steps_per_second": 3.398,
+      "step": 4220
+    },
+    {
+      "epoch": 11.0,
+      "grad_norm": 0.663550078868866,
+      "learning_rate": 4.771464646464647e-05,
+      "loss": 4.1418,
+      "step": 4642
+    },
+    {
+      "epoch": 11.0,
+      "eval_loss": 4.379039764404297,
+      "eval_runtime": 29.9854,
+      "eval_samples_per_second": 918.515,
+      "eval_steps_per_second": 3.602,
+      "step": 4642
+    },
+    {
+      "epoch": 12.0,
+      "grad_norm": 0.744036078453064,
+      "learning_rate": 4.238636363636364e-05,
+      "loss": 4.1029,
+      "step": 5064
+    },
+    {
+      "epoch": 12.0,
+      "eval_loss": 4.383896350860596,
+      "eval_runtime": 25.563,
+      "eval_samples_per_second": 1077.418,
+      "eval_steps_per_second": 4.225,
+      "step": 5064
+    },
+    {
+      "epoch": 13.0,
+      "grad_norm": 0.8011893630027771,
+      "learning_rate": 3.705808080808081e-05,
+      "loss": 4.0643,
+      "step": 5486
+    },
+    {
+      "epoch": 13.0,
+      "eval_loss": 4.391408920288086,
+      "eval_runtime": 30.7558,
+      "eval_samples_per_second": 895.506,
+      "eval_steps_per_second": 3.512,
+      "step": 5486
+    },
+    {
+      "epoch": 14.0,
+      "grad_norm": 0.8544487357139587,
+      "learning_rate": 3.172979797979798e-05,
+      "loss": 4.0259,
+      "step": 5908
+    },
+    {
+      "epoch": 14.0,
+      "eval_loss": 4.40255880355835,
+      "eval_runtime": 31.2899,
+      "eval_samples_per_second": 880.219,
+      "eval_steps_per_second": 3.452,
+      "step": 5908
+    },
+    {
+      "epoch": 15.0,
+      "grad_norm": 0.9503664970397949,
+      "learning_rate": 2.6401515151515155e-05,
+      "loss": 3.9896,
+      "step": 6330
+    },
+    {
+      "epoch": 15.0,
+      "eval_loss": 4.4131035804748535,
+      "eval_runtime": 21.2356,
+      "eval_samples_per_second": 1296.974,
+      "eval_steps_per_second": 5.086,
+      "step": 6330
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 8440,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 20,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.0868866670592e+17,
+  "train_batch_size": 256,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-6500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e427d3053d40db71f4c8bd911dbb23cf2c24d836bdfda62c5985c167b6b35982
+size 5969