Training in progress, step 16000, checkpoint

checkpoint-16000/added_tokens.json

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

checkpoint-16000/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-16000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:acbb336c7618a1676609bd7522573a01877f5a8d36e5a5e10f402a3de6b00a88
+size 592141189

checkpoint-16000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7216ba8b8aa19e0591d4c0a066a69d3d912d7246e225ae22e796c9a5642dec2
+size 296072722

checkpoint-16000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e900211f8f9def3755a0198b3c0818d577309a05eaea9641876e33a2a0509c0e
+size 14575

checkpoint-16000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ff3978389d73985f5b74f24f9d23c58ca2f8f60e636f62bd9e974071be339f8
+size 627

checkpoint-16000/special_tokens_map.json

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

checkpoint-16000/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-16000/trainer_state.json

@@ -0,0 +1,265 @@
+{
+  "best_metric": 1.0855979919433594,
+  "best_model_checkpoint": "/kaggle/output/checkpoint-16000",
+  "epoch": 0.651890482398957,
+  "eval_steps": 1000,
+  "global_step": 16000,
+  "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": 1.05,
+      "step": 1
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 2.7750000000000004e-08,
+      "loss": 1.1403,
+      "step": 1000
+    },
+    {
+      "epoch": 0.04,
+      "eval_accuracy": 0.33712574850299404,
+      "eval_loss": 1.1030837297439575,
+      "eval_runtime": 29.7882,
+      "eval_samples_per_second": 168.187,
+      "eval_steps_per_second": 21.049,
+      "step": 1000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 5.5527777777777784e-08,
+      "loss": 1.1194,
+      "step": 2000
+    },
+    {
+      "epoch": 0.08,
+      "eval_accuracy": 0.34331337325349304,
+      "eval_loss": 1.1002165079116821,
+      "eval_runtime": 29.8685,
+      "eval_samples_per_second": 167.735,
+      "eval_steps_per_second": 20.992,
+      "step": 2000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 8.327777777777778e-08,
+      "loss": 1.1194,
+      "step": 3000
+    },
+    {
+      "epoch": 0.12,
+      "eval_accuracy": 0.3471057884231537,
+      "eval_loss": 1.101266622543335,
+      "eval_runtime": 29.709,
+      "eval_samples_per_second": 168.636,
+      "eval_steps_per_second": 21.105,
+      "step": 3000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.1105555555555557e-07,
+      "loss": 1.1166,
+      "step": 4000
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.34311377245508984,
+      "eval_loss": 1.0984796285629272,
+      "eval_runtime": 29.9401,
+      "eval_samples_per_second": 167.334,
+      "eval_steps_per_second": 20.942,
+      "step": 4000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.3883333333333335e-07,
+      "loss": 1.1137,
+      "step": 5000
+    },
+    {
+      "epoch": 0.2,
+      "eval_accuracy": 0.3403193612774451,
+      "eval_loss": 1.0974279642105103,
+      "eval_runtime": 29.8185,
+      "eval_samples_per_second": 168.016,
+      "eval_steps_per_second": 21.027,
+      "step": 5000
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.6658333333333335e-07,
+      "loss": 1.1109,
+      "step": 6000
+    },
+    {
+      "epoch": 0.24,
+      "eval_accuracy": 0.3592814371257485,
+      "eval_loss": 1.0984892845153809,
+      "eval_runtime": 29.893,
+      "eval_samples_per_second": 167.598,
+      "eval_steps_per_second": 20.975,
+      "step": 6000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9436111111111112e-07,
+      "loss": 1.1122,
+      "step": 7000
+    },
+    {
+      "epoch": 0.29,
+      "eval_accuracy": 0.36327345309381237,
+      "eval_loss": 1.0960686206817627,
+      "eval_runtime": 29.8315,
+      "eval_samples_per_second": 167.943,
+      "eval_steps_per_second": 21.018,
+      "step": 7000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 2.2213888888888891e-07,
+      "loss": 1.1091,
+      "step": 8000
+    },
+    {
+      "epoch": 0.33,
+      "eval_accuracy": 0.35728542914171657,
+      "eval_loss": 1.0958523750305176,
+      "eval_runtime": 29.9494,
+      "eval_samples_per_second": 167.282,
+      "eval_steps_per_second": 20.935,
+      "step": 8000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 2.4988888888888893e-07,
+      "loss": 1.1077,
+      "step": 9000
+    },
+    {
+      "epoch": 0.37,
+      "eval_accuracy": 0.35528942115768464,
+      "eval_loss": 1.1003851890563965,
+      "eval_runtime": 29.9762,
+      "eval_samples_per_second": 167.133,
+      "eval_steps_per_second": 20.917,
+      "step": 9000
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.776666666666667e-07,
+      "loss": 1.1091,
+      "step": 10000
+    },
+    {
+      "epoch": 0.41,
+      "eval_accuracy": 0.3790419161676647,
+      "eval_loss": 1.0951762199401855,
+      "eval_runtime": 29.8593,
+      "eval_samples_per_second": 167.787,
+      "eval_steps_per_second": 20.998,
+      "step": 10000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 3.0541666666666667e-07,
+      "loss": 1.1047,
+      "step": 11000
+    },
+    {
+      "epoch": 0.45,
+      "eval_accuracy": 0.3780439121756487,
+      "eval_loss": 1.0928910970687866,
+      "eval_runtime": 29.8768,
+      "eval_samples_per_second": 167.689,
+      "eval_steps_per_second": 20.986,
+      "step": 11000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 3.3319444444444444e-07,
+      "loss": 1.1053,
+      "step": 12000
+    },
+    {
+      "epoch": 0.49,
+      "eval_accuracy": 0.38323353293413176,
+      "eval_loss": 1.0910215377807617,
+      "eval_runtime": 29.8703,
+      "eval_samples_per_second": 167.725,
+      "eval_steps_per_second": 20.991,
+      "step": 12000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 3.6094444444444446e-07,
+      "loss": 1.1035,
+      "step": 13000
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.34850299401197604,
+      "eval_loss": 1.0930161476135254,
+      "eval_runtime": 29.8865,
+      "eval_samples_per_second": 167.634,
+      "eval_steps_per_second": 20.979,
+      "step": 13000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 3.8872222222222223e-07,
+      "loss": 1.1002,
+      "step": 14000
+    },
+    {
+      "epoch": 0.57,
+      "eval_accuracy": 0.3646706586826347,
+      "eval_loss": 1.093497633934021,
+      "eval_runtime": 30.0222,
+      "eval_samples_per_second": 166.877,
+      "eval_steps_per_second": 20.885,
+      "step": 14000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.1647222222222225e-07,
+      "loss": 1.1025,
+      "step": 15000
+    },
+    {
+      "epoch": 0.61,
+      "eval_accuracy": 0.37924151696606784,
+      "eval_loss": 1.0900779962539673,
+      "eval_runtime": 30.0049,
+      "eval_samples_per_second": 166.973,
+      "eval_steps_per_second": 20.897,
+      "step": 15000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.4425e-07,
+      "loss": 1.0992,
+      "step": 16000
+    },
+    {
+      "epoch": 0.65,
+      "eval_accuracy": 0.40119760479041916,
+      "eval_loss": 1.0855979919433594,
+      "eval_runtime": 30.1095,
+      "eval_samples_per_second": 166.393,
+      "eval_steps_per_second": 20.824,
+      "step": 16000
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 10000000,
+  "num_train_epochs": 408,
+  "save_steps": 1000,
+  "total_flos": 1.6723294027776e+16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-16000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40a0dfc08574fef5341c0802f4b4b8be93cd36a17a96b8a7b07271ea753deed4
+size 4027