Training in progress, step 17000, checkpoint

checkpoint-17000/added_tokens.json

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

checkpoint-17000/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": 6,
+  "n_layer": 12,
+  "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": "cls_index",
+  "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-17000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1dc42e2a80665a6d547b050f5753d0a7b0be6559d944e3345e76061fa6cf0946
+size 932401285

checkpoint-17000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30b499b008bad2c433d855fb32b2af541b4fe2d73bdf913d84a3e082f476c94b
+size 466205145

checkpoint-17000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37edc7580ae044ea11563bbeacd5cb72dcb4da5e947e3a698584768d7a6e1141
+size 14575

checkpoint-17000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:238b5e087a1203f28b0cc7cb0b8ce535844175bb05bd8859ee086ca4f463974a
+size 627

checkpoint-17000/special_tokens_map.json

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

checkpoint-17000/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-17000/trainer_state.json

@@ -0,0 +1,280 @@
+{
+  "best_metric": 1.0795228481292725,
+  "best_model_checkpoint": "/kaggle/output/checkpoint-17000",
+  "epoch": 0.6926336375488917,
+  "eval_steps": 1000,
+  "global_step": 17000,
+  "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": 1.2076,
+      "step": 1
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 2.7750000000000004e-08,
+      "loss": 1.1301,
+      "step": 1000
+    },
+    {
+      "epoch": 0.04,
+      "eval_accuracy": 0.3321357285429142,
+      "eval_loss": 1.1047232151031494,
+      "eval_runtime": 50.8344,
+      "eval_samples_per_second": 98.555,
+      "eval_steps_per_second": 12.334,
+      "step": 1000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 5.5527777777777784e-08,
+      "loss": 1.1286,
+      "step": 2000
+    },
+    {
+      "epoch": 0.08,
+      "eval_accuracy": 0.3377245508982036,
+      "eval_loss": 1.1038333177566528,
+      "eval_runtime": 50.8114,
+      "eval_samples_per_second": 98.6,
+      "eval_steps_per_second": 12.34,
+      "step": 2000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 8.327777777777778e-08,
+      "loss": 1.1214,
+      "step": 3000
+    },
+    {
+      "epoch": 0.12,
+      "eval_accuracy": 0.3409181636726547,
+      "eval_loss": 1.103529930114746,
+      "eval_runtime": 50.8258,
+      "eval_samples_per_second": 98.572,
+      "eval_steps_per_second": 12.336,
+      "step": 3000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.1105555555555557e-07,
+      "loss": 1.1249,
+      "step": 4000
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.3469061876247505,
+      "eval_loss": 1.1000434160232544,
+      "eval_runtime": 50.6414,
+      "eval_samples_per_second": 98.931,
+      "eval_steps_per_second": 12.381,
+      "step": 4000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.3880555555555558e-07,
+      "loss": 1.1206,
+      "step": 5000
+    },
+    {
+      "epoch": 0.2,
+      "eval_accuracy": 0.34171656686626745,
+      "eval_loss": 1.0996097326278687,
+      "eval_runtime": 50.6956,
+      "eval_samples_per_second": 98.825,
+      "eval_steps_per_second": 12.368,
+      "step": 5000
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.6658333333333335e-07,
+      "loss": 1.1161,
+      "step": 6000
+    },
+    {
+      "epoch": 0.24,
+      "eval_accuracy": 0.3590818363273453,
+      "eval_loss": 1.096977710723877,
+      "eval_runtime": 50.5826,
+      "eval_samples_per_second": 99.046,
+      "eval_steps_per_second": 12.396,
+      "step": 6000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9433333333333334e-07,
+      "loss": 1.1138,
+      "step": 7000
+    },
+    {
+      "epoch": 0.29,
+      "eval_accuracy": 0.3532934131736527,
+      "eval_loss": 1.0949815511703491,
+      "eval_runtime": 50.5604,
+      "eval_samples_per_second": 99.089,
+      "eval_steps_per_second": 12.401,
+      "step": 7000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 2.2211111111111114e-07,
+      "loss": 1.1135,
+      "step": 8000
+    },
+    {
+      "epoch": 0.33,
+      "eval_accuracy": 0.36327345309381237,
+      "eval_loss": 1.093558430671692,
+      "eval_runtime": 50.608,
+      "eval_samples_per_second": 98.996,
+      "eval_steps_per_second": 12.389,
+      "step": 8000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 2.4986111111111113e-07,
+      "loss": 1.1063,
+      "step": 9000
+    },
+    {
+      "epoch": 0.37,
+      "eval_accuracy": 0.35548902195608784,
+      "eval_loss": 1.0992107391357422,
+      "eval_runtime": 50.5434,
+      "eval_samples_per_second": 99.123,
+      "eval_steps_per_second": 12.405,
+      "step": 9000
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.776388888888889e-07,
+      "loss": 1.1083,
+      "step": 10000
+    },
+    {
+      "epoch": 0.41,
+      "eval_accuracy": 0.3836327345309381,
+      "eval_loss": 1.0920206308364868,
+      "eval_runtime": 50.5559,
+      "eval_samples_per_second": 99.098,
+      "eval_steps_per_second": 12.402,
+      "step": 10000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 3.053888888888889e-07,
+      "loss": 1.1044,
+      "step": 11000
+    },
+    {
+      "epoch": 0.45,
+      "eval_accuracy": 0.38762475049900197,
+      "eval_loss": 1.0904459953308105,
+      "eval_runtime": 50.9725,
+      "eval_samples_per_second": 98.288,
+      "eval_steps_per_second": 12.301,
+      "step": 11000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 3.331666666666667e-07,
+      "loss": 1.1019,
+      "step": 12000
+    },
+    {
+      "epoch": 0.49,
+      "eval_accuracy": 0.3912175648702595,
+      "eval_loss": 1.088543176651001,
+      "eval_runtime": 50.8297,
+      "eval_samples_per_second": 98.564,
+      "eval_steps_per_second": 12.335,
+      "step": 12000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 3.609166666666667e-07,
+      "loss": 1.1032,
+      "step": 13000
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.3435129740518962,
+      "eval_loss": 1.0940682888031006,
+      "eval_runtime": 51.0031,
+      "eval_samples_per_second": 98.229,
+      "eval_steps_per_second": 12.293,
+      "step": 13000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 3.886944444444445e-07,
+      "loss": 1.1006,
+      "step": 14000
+    },
+    {
+      "epoch": 0.57,
+      "eval_accuracy": 0.36826347305389223,
+      "eval_loss": 1.0924910306930542,
+      "eval_runtime": 51.2753,
+      "eval_samples_per_second": 97.708,
+      "eval_steps_per_second": 12.228,
+      "step": 14000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.164444444444445e-07,
+      "loss": 1.1,
+      "step": 15000
+    },
+    {
+      "epoch": 0.61,
+      "eval_accuracy": 0.38323353293413176,
+      "eval_loss": 1.0879395008087158,
+      "eval_runtime": 51.3783,
+      "eval_samples_per_second": 97.512,
+      "eval_steps_per_second": 12.204,
+      "step": 15000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.442222222222223e-07,
+      "loss": 1.0982,
+      "step": 16000
+    },
+    {
+      "epoch": 0.65,
+      "eval_accuracy": 0.3964071856287425,
+      "eval_loss": 1.0822893381118774,
+      "eval_runtime": 51.2346,
+      "eval_samples_per_second": 97.786,
+      "eval_steps_per_second": 12.238,
+      "step": 16000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.7197222222222224e-07,
+      "loss": 1.0953,
+      "step": 17000
+    },
+    {
+      "epoch": 0.69,
+      "eval_accuracy": 0.4121756487025948,
+      "eval_loss": 1.0795228481292725,
+      "eval_runtime": 51.31,
+      "eval_samples_per_second": 97.642,
+      "eval_steps_per_second": 12.22,
+      "step": 17000
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 10000000,
+  "num_train_epochs": 408,
+  "save_steps": 1000,
+  "total_flos": 3.5536037216256e+16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-17000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:933e5c21c45262af343f5d708824cb21f8d339a9e2662221f8b52a4157df35fa
+size 4091