Training in progress, step 20000, checkpoint

checkpoint-20000/added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<bos>": 40478,
+  "<eos>": 40479,
+  "<pad>": 40480
+}

checkpoint-20000/config.json

@@ -0,0 +1,46 @@
+{
+  "_name_or_path": "openai-gpt",
+  "afn": "gelu",
+  "architectures": [
+    "OpenAIGPTForSequenceClassification"
+  ],
+  "attn_pdrop": 0.1,
+  "embd_pdrop": 0.1,
+  "id2label": {
+    "0": "entailment",
+    "1": "neutral",
+    "2": "contradiction"
+  },
+  "initializer_range": 0.02,
+  "label2id": {
+    "contradiction": 2,
+    "entailment": 0,
+    "neutral": 1
+  },
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "openai-gpt",
+  "n_ctx": 512,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_layer": 6,
+  "n_positions": 512,
+  "n_special": 0,
+  "pad_token_id": 40480,
+  "predict_special_tokens": true,
+  "problem_type": "single_label_classification",
+  "resid_pdrop": 0.1,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "last",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "torch_dtype": "float32",
+  "transformers_version": "4.33.0.dev0",
+  "vocab_size": 40481
+}

checkpoint-20000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc5ec1c8f137466a39a4a90ead4226d08a65aa070768c7f9914bfa1599bac724
+size 592141189

checkpoint-20000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e482e28536b0fd6c52c2059e0b9b31ab6875c83551feb3d1e2e33c467582f283
+size 296072722

checkpoint-20000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e67006fdb09cea4275875517b1a9922da48d611fbbfb2bbdb1e53446bc4f6af2
+size 14575

checkpoint-20000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81ca0953acb5d3b48ae21da5bf1c9229a3f9adae8b44088f07afb7e0ef95196a
+size 627

checkpoint-20000/special_tokens_map.json

@@ -0,0 +1,6 @@
+{
+  "bos_token": "<bos>",
+  "eos_token": "<eos>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

checkpoint-20000/tokenizer_config.json

@@ -0,0 +1,7 @@
+{
+  "clean_up_tokenization_spaces": true,
+  "model_max_length": 512,
+  "padding_side": "left",
+  "tokenizer_class": "OpenAIGPTTokenizer",
+  "unk_token": "<unk>"
+}

checkpoint-20000/trainer_state.json

@@ -0,0 +1,325 @@
+{
+  "best_metric": 1.0817394256591797,
+  "best_model_checkpoint": "/kaggle/output/checkpoint-20000",
+  "epoch": 0.8148631029986962,
+  "eval_steps": 1000,
+  "global_step": 20000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0,
+      "learning_rate": 2.7777777777777777e-11,
+      "loss": 1.2358,
+      "step": 1
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 2.7750000000000004e-08,
+      "loss": 1.1519,
+      "step": 1000
+    },
+    {
+      "epoch": 0.04,
+      "eval_accuracy": 0.3201596806387226,
+      "eval_loss": 1.1120171546936035,
+      "eval_runtime": 29.027,
+      "eval_samples_per_second": 172.598,
+      "eval_steps_per_second": 21.601,
+      "step": 1000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 5.5527777777777784e-08,
+      "loss": 1.128,
+      "step": 2000
+    },
+    {
+      "epoch": 0.08,
+      "eval_accuracy": 0.31796407185628744,
+      "eval_loss": 1.108055830001831,
+      "eval_runtime": 28.9944,
+      "eval_samples_per_second": 172.792,
+      "eval_steps_per_second": 21.625,
+      "step": 2000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 8.327777777777778e-08,
+      "loss": 1.122,
+      "step": 3000
+    },
+    {
+      "epoch": 0.12,
+      "eval_accuracy": 0.3345309381237525,
+      "eval_loss": 1.1074285507202148,
+      "eval_runtime": 28.9901,
+      "eval_samples_per_second": 172.818,
+      "eval_steps_per_second": 21.628,
+      "step": 3000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.1105555555555557e-07,
+      "loss": 1.1196,
+      "step": 4000
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.3337325349301397,
+      "eval_loss": 1.1045424938201904,
+      "eval_runtime": 28.9791,
+      "eval_samples_per_second": 172.883,
+      "eval_steps_per_second": 21.636,
+      "step": 4000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.3880555555555558e-07,
+      "loss": 1.1169,
+      "step": 5000
+    },
+    {
+      "epoch": 0.2,
+      "eval_accuracy": 0.3397205588822355,
+      "eval_loss": 1.1033179759979248,
+      "eval_runtime": 29.0295,
+      "eval_samples_per_second": 172.583,
+      "eval_steps_per_second": 21.599,
+      "step": 5000
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.6658333333333335e-07,
+      "loss": 1.118,
+      "step": 6000
+    },
+    {
+      "epoch": 0.24,
+      "eval_accuracy": 0.3401197604790419,
+      "eval_loss": 1.1030455827713013,
+      "eval_runtime": 28.9748,
+      "eval_samples_per_second": 172.909,
+      "eval_steps_per_second": 21.639,
+      "step": 6000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9433333333333334e-07,
+      "loss": 1.1135,
+      "step": 7000
+    },
+    {
+      "epoch": 0.29,
+      "eval_accuracy": 0.35568862275449104,
+      "eval_loss": 1.0998502969741821,
+      "eval_runtime": 28.9964,
+      "eval_samples_per_second": 172.78,
+      "eval_steps_per_second": 21.623,
+      "step": 7000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 2.2211111111111114e-07,
+      "loss": 1.1136,
+      "step": 8000
+    },
+    {
+      "epoch": 0.33,
+      "eval_accuracy": 0.34510978043912177,
+      "eval_loss": 1.098555564880371,
+      "eval_runtime": 28.983,
+      "eval_samples_per_second": 172.86,
+      "eval_steps_per_second": 21.633,
+      "step": 8000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 2.4986111111111113e-07,
+      "loss": 1.11,
+      "step": 9000
+    },
+    {
+      "epoch": 0.37,
+      "eval_accuracy": 0.3411177644710579,
+      "eval_loss": 1.1035741567611694,
+      "eval_runtime": 28.9758,
+      "eval_samples_per_second": 172.903,
+      "eval_steps_per_second": 21.639,
+      "step": 9000
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.776388888888889e-07,
+      "loss": 1.1076,
+      "step": 10000
+    },
+    {
+      "epoch": 0.41,
+      "eval_accuracy": 0.35788423153692617,
+      "eval_loss": 1.0970662832260132,
+      "eval_runtime": 29.0262,
+      "eval_samples_per_second": 172.603,
+      "eval_steps_per_second": 21.601,
+      "step": 10000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 3.0541666666666667e-07,
+      "loss": 1.1067,
+      "step": 11000
+    },
+    {
+      "epoch": 0.45,
+      "eval_accuracy": 0.37425149700598803,
+      "eval_loss": 1.094608187675476,
+      "eval_runtime": 28.8738,
+      "eval_samples_per_second": 173.513,
+      "eval_steps_per_second": 21.715,
+      "step": 11000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 3.3319444444444444e-07,
+      "loss": 1.1032,
+      "step": 12000
+    },
+    {
+      "epoch": 0.49,
+      "eval_accuracy": 0.37924151696606784,
+      "eval_loss": 1.092067837715149,
+      "eval_runtime": 28.9514,
+      "eval_samples_per_second": 173.049,
+      "eval_steps_per_second": 21.657,
+      "step": 12000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 3.6094444444444446e-07,
+      "loss": 1.1052,
+      "step": 13000
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.3435129740518962,
+      "eval_loss": 1.0962508916854858,
+      "eval_runtime": 29.2203,
+      "eval_samples_per_second": 171.456,
+      "eval_steps_per_second": 21.458,
+      "step": 13000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 3.8872222222222223e-07,
+      "loss": 1.1015,
+      "step": 14000
+    },
+    {
+      "epoch": 0.57,
+      "eval_accuracy": 0.36806387225548903,
+      "eval_loss": 1.0941067934036255,
+      "eval_runtime": 29.2088,
+      "eval_samples_per_second": 171.524,
+      "eval_steps_per_second": 21.466,
+      "step": 14000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.1650000000000006e-07,
+      "loss": 1.0986,
+      "step": 15000
+    },
+    {
+      "epoch": 0.61,
+      "eval_accuracy": 0.3718562874251497,
+      "eval_loss": 1.0906143188476562,
+      "eval_runtime": 28.9463,
+      "eval_samples_per_second": 173.079,
+      "eval_steps_per_second": 21.661,
+      "step": 15000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.4425e-07,
+      "loss": 1.0984,
+      "step": 16000
+    },
+    {
+      "epoch": 0.65,
+      "eval_accuracy": 0.38183632734530937,
+      "eval_loss": 1.0862348079681396,
+      "eval_runtime": 28.9909,
+      "eval_samples_per_second": 172.813,
+      "eval_steps_per_second": 21.627,
+      "step": 16000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.7202777777777785e-07,
+      "loss": 1.0959,
+      "step": 17000
+    },
+    {
+      "epoch": 0.69,
+      "eval_accuracy": 0.40079840319361276,
+      "eval_loss": 1.0846956968307495,
+      "eval_runtime": 28.9606,
+      "eval_samples_per_second": 172.994,
+      "eval_steps_per_second": 21.65,
+      "step": 17000
+    },
+    {
+      "epoch": 0.73,
+      "learning_rate": 4.998055555555556e-07,
+      "loss": 1.0965,
+      "step": 18000
+    },
+    {
+      "epoch": 0.73,
+      "eval_accuracy": 0.4043912175648703,
+      "eval_loss": 1.086153507232666,
+      "eval_runtime": 28.8411,
+      "eval_samples_per_second": 173.71,
+      "eval_steps_per_second": 21.74,
+      "step": 18000
+    },
+    {
+      "epoch": 0.77,
+      "learning_rate": 5.275555555555556e-07,
+      "loss": 1.0966,
+      "step": 19000
+    },
+    {
+      "epoch": 0.77,
+      "eval_accuracy": 0.4041916167664671,
+      "eval_loss": 1.0824861526489258,
+      "eval_runtime": 28.9652,
+      "eval_samples_per_second": 172.966,
+      "eval_steps_per_second": 21.647,
+      "step": 19000
+    },
+    {
+      "epoch": 0.81,
+      "learning_rate": 5.553333333333334e-07,
+      "loss": 1.0943,
+      "step": 20000
+    },
+    {
+      "epoch": 0.81,
+      "eval_accuracy": 0.4103792415169661,
+      "eval_loss": 1.0817394256591797,
+      "eval_runtime": 28.9548,
+      "eval_samples_per_second": 173.028,
+      "eval_steps_per_second": 21.654,
+      "step": 20000
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 10000000,
+  "num_train_epochs": 408,
+  "save_steps": 1000,
+  "total_flos": 2.090411753472e+16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-20000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35131e4915b0e8d5c12cfe76344eb3f82740d8fb21233ad6a425c6f5e63a567e
+size 4027