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": 12,
+  "n_layer": 4,
+  "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-17000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e901d32d06af54f34f37f1e20516d6eb6d2be9dd6bf5c18d55463018c20fe001
+size 478721221

checkpoint-17000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74ae98b9217e29af581aafd208035c9cdd538d8521a0ab2d53832695a2d29d6e
+size 239361994

checkpoint-17000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef3b7a62f4b6375332ef22858bb9b6c9a6c0f243727c23efd075dc5ef78d318f
+size 14575

checkpoint-17000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d76441294bf83cbcd53c2c0e413890a98c07a97df86111ab308281c3f73d709e
+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": 1.0791982412338257,
+  "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": 1.2184,
+      "step": 1
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 2.7750000000000004e-08,
+      "loss": 1.1394,
+      "step": 1000
+    },
+    {
+      "epoch": 0.04,
+      "eval_accuracy": 0.3327345309381238,
+      "eval_loss": 1.1149410009384155,
+      "eval_runtime": 20.6803,
+      "eval_samples_per_second": 242.26,
+      "eval_steps_per_second": 30.319,
+      "step": 1000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 5.5527777777777784e-08,
+      "loss": 1.1141,
+      "step": 2000
+    },
+    {
+      "epoch": 0.08,
+      "eval_accuracy": 0.3401197604790419,
+      "eval_loss": 1.104099988937378,
+      "eval_runtime": 20.8477,
+      "eval_samples_per_second": 240.314,
+      "eval_steps_per_second": 30.075,
+      "step": 2000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 8.330555555555556e-08,
+      "loss": 1.116,
+      "step": 3000
+    },
+    {
+      "epoch": 0.12,
+      "eval_accuracy": 0.3407185628742515,
+      "eval_loss": 1.1040862798690796,
+      "eval_runtime": 20.6818,
+      "eval_samples_per_second": 242.242,
+      "eval_steps_per_second": 30.317,
+      "step": 3000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.1108333333333333e-07,
+      "loss": 1.1158,
+      "step": 4000
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.32894211576846305,
+      "eval_loss": 1.1020556688308716,
+      "eval_runtime": 20.8541,
+      "eval_samples_per_second": 240.241,
+      "eval_steps_per_second": 30.066,
+      "step": 4000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.3883333333333335e-07,
+      "loss": 1.1135,
+      "step": 5000
+    },
+    {
+      "epoch": 0.2,
+      "eval_accuracy": 0.34271457085828344,
+      "eval_loss": 1.1008552312850952,
+      "eval_runtime": 20.8055,
+      "eval_samples_per_second": 240.802,
+      "eval_steps_per_second": 30.136,
+      "step": 5000
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.6658333333333335e-07,
+      "loss": 1.1121,
+      "step": 6000
+    },
+    {
+      "epoch": 0.24,
+      "eval_accuracy": 0.3395209580838323,
+      "eval_loss": 1.1004050970077515,
+      "eval_runtime": 20.8985,
+      "eval_samples_per_second": 239.731,
+      "eval_steps_per_second": 30.002,
+      "step": 6000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9436111111111112e-07,
+      "loss": 1.1089,
+      "step": 7000
+    },
+    {
+      "epoch": 0.29,
+      "eval_accuracy": 0.35788423153692617,
+      "eval_loss": 1.0985721349716187,
+      "eval_runtime": 20.84,
+      "eval_samples_per_second": 240.403,
+      "eval_steps_per_second": 30.086,
+      "step": 7000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 2.2213888888888891e-07,
+      "loss": 1.1079,
+      "step": 8000
+    },
+    {
+      "epoch": 0.33,
+      "eval_accuracy": 0.3331337325349301,
+      "eval_loss": 1.098374843597412,
+      "eval_runtime": 20.7886,
+      "eval_samples_per_second": 240.998,
+      "eval_steps_per_second": 30.161,
+      "step": 8000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 2.4988888888888893e-07,
+      "loss": 1.1087,
+      "step": 9000
+    },
+    {
+      "epoch": 0.37,
+      "eval_accuracy": 0.34510978043912177,
+      "eval_loss": 1.0993521213531494,
+      "eval_runtime": 20.782,
+      "eval_samples_per_second": 241.074,
+      "eval_steps_per_second": 30.17,
+      "step": 9000
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.776666666666667e-07,
+      "loss": 1.109,
+      "step": 10000
+    },
+    {
+      "epoch": 0.41,
+      "eval_accuracy": 0.3475049900199601,
+      "eval_loss": 1.0967597961425781,
+      "eval_runtime": 20.6798,
+      "eval_samples_per_second": 242.265,
+      "eval_steps_per_second": 30.319,
+      "step": 10000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 3.054444444444444e-07,
+      "loss": 1.1052,
+      "step": 11000
+    },
+    {
+      "epoch": 0.45,
+      "eval_accuracy": 0.37544910179640717,
+      "eval_loss": 1.0941349267959595,
+      "eval_runtime": 20.8641,
+      "eval_samples_per_second": 240.126,
+      "eval_steps_per_second": 30.052,
+      "step": 11000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 3.3322222222222225e-07,
+      "loss": 1.105,
+      "step": 12000
+    },
+    {
+      "epoch": 0.49,
+      "eval_accuracy": 0.3834331337325349,
+      "eval_loss": 1.0927647352218628,
+      "eval_runtime": 20.6541,
+      "eval_samples_per_second": 242.567,
+      "eval_steps_per_second": 30.357,
+      "step": 12000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 3.609722222222222e-07,
+      "loss": 1.1016,
+      "step": 13000
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.3457085828343313,
+      "eval_loss": 1.0942081212997437,
+      "eval_runtime": 21.0733,
+      "eval_samples_per_second": 237.742,
+      "eval_steps_per_second": 29.753,
+      "step": 13000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 3.8875e-07,
+      "loss": 1.1031,
+      "step": 14000
+    },
+    {
+      "epoch": 0.57,
+      "eval_accuracy": 0.37005988023952097,
+      "eval_loss": 1.0918152332305908,
+      "eval_runtime": 20.9151,
+      "eval_samples_per_second": 239.54,
+      "eval_steps_per_second": 29.978,
+      "step": 14000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.1652777777777786e-07,
+      "loss": 1.1026,
+      "step": 15000
+    },
+    {
+      "epoch": 0.61,
+      "eval_accuracy": 0.3790419161676647,
+      "eval_loss": 1.0895211696624756,
+      "eval_runtime": 21.0591,
+      "eval_samples_per_second": 237.902,
+      "eval_steps_per_second": 29.773,
+      "step": 15000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.4427777777777783e-07,
+      "loss": 1.0988,
+      "step": 16000
+    },
+    {
+      "epoch": 0.65,
+      "eval_accuracy": 0.4101796407185629,
+      "eval_loss": 1.0852997303009033,
+      "eval_runtime": 20.9509,
+      "eval_samples_per_second": 239.131,
+      "eval_steps_per_second": 29.927,
+      "step": 16000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.720555555555556e-07,
+      "loss": 1.0974,
+      "step": 17000
+    },
+    {
+      "epoch": 0.69,
+      "eval_accuracy": 0.43213572854291415,
+      "eval_loss": 1.0791982412338257,
+      "eval_runtime": 20.7526,
+      "eval_samples_per_second": 241.415,
+      "eval_steps_per_second": 30.213,
+      "step": 17000
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 10000000,
+  "num_train_epochs": 408,
+  "save_steps": 1000,
+  "total_flos": 1.1845987467264e+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:d42437e01c6a8e691de275f2b2ef4e9ee68d60c829bca2508adb832bd55170f9
+size 4027