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": 2,
+  "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 sha256:33c60fa3853a2dcc8450d4812dc0362ea23de80645dcfae434077d5a20bab683
+size 365301235

checkpoint-17000/pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4ce0e5bda2d0c5845f1264c99626e915b45954a3a53d6e46285c96fe4ae4dafb
+size 182651202

checkpoint-17000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cff85fc14d0125cd79ba38bc3fa2a09954c81b5e656428686969b58cec7b229d
+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": 1.0901482105255127,
+  "best_model_checkpoint": "/kaggle/output/checkpoint-16000",
+  "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": 1.1078,
+      "step": 1
+    },
+    {
+      "epoch": 0.04,
+      "learning_rate": 2.7750000000000004e-08,
+      "loss": 1.128,
+      "step": 1000
+    },
+    {
+      "epoch": 0.04,
+      "eval_accuracy": 0.330938123752495,
+      "eval_loss": 1.1083118915557861,
+      "eval_runtime": 12.3979,
+      "eval_samples_per_second": 404.1,
+      "eval_steps_per_second": 50.573,
+      "step": 1000
+    },
+    {
+      "epoch": 0.08,
+      "learning_rate": 5.5527777777777784e-08,
+      "loss": 1.1164,
+      "step": 2000
+    },
+    {
+      "epoch": 0.08,
+      "eval_accuracy": 0.3323353293413174,
+      "eval_loss": 1.102672815322876,
+      "eval_runtime": 12.2861,
+      "eval_samples_per_second": 407.778,
+      "eval_steps_per_second": 51.033,
+      "step": 2000
+    },
+    {
+      "epoch": 0.12,
+      "learning_rate": 8.327777777777778e-08,
+      "loss": 1.1158,
+      "step": 3000
+    },
+    {
+      "epoch": 0.12,
+      "eval_accuracy": 0.34211576846307384,
+      "eval_loss": 1.1024774312973022,
+      "eval_runtime": 12.3746,
+      "eval_samples_per_second": 404.863,
+      "eval_steps_per_second": 50.668,
+      "step": 3000
+    },
+    {
+      "epoch": 0.16,
+      "learning_rate": 1.1105555555555557e-07,
+      "loss": 1.1115,
+      "step": 4000
+    },
+    {
+      "epoch": 0.16,
+      "eval_accuracy": 0.34890219560878244,
+      "eval_loss": 1.1012382507324219,
+      "eval_runtime": 12.5405,
+      "eval_samples_per_second": 399.505,
+      "eval_steps_per_second": 49.998,
+      "step": 4000
+    },
+    {
+      "epoch": 0.2,
+      "learning_rate": 1.3880555555555558e-07,
+      "loss": 1.1141,
+      "step": 5000
+    },
+    {
+      "epoch": 0.2,
+      "eval_accuracy": 0.3534930139720559,
+      "eval_loss": 1.1002192497253418,
+      "eval_runtime": 12.3001,
+      "eval_samples_per_second": 407.314,
+      "eval_steps_per_second": 50.975,
+      "step": 5000
+    },
+    {
+      "epoch": 0.24,
+      "learning_rate": 1.6658333333333335e-07,
+      "loss": 1.112,
+      "step": 6000
+    },
+    {
+      "epoch": 0.24,
+      "eval_accuracy": 0.3524950099800399,
+      "eval_loss": 1.100056767463684,
+      "eval_runtime": 12.4062,
+      "eval_samples_per_second": 403.831,
+      "eval_steps_per_second": 50.539,
+      "step": 6000
+    },
+    {
+      "epoch": 0.29,
+      "learning_rate": 1.9433333333333334e-07,
+      "loss": 1.1115,
+      "step": 7000
+    },
+    {
+      "epoch": 0.29,
+      "eval_accuracy": 0.34331337325349304,
+      "eval_loss": 1.098986029624939,
+      "eval_runtime": 12.4456,
+      "eval_samples_per_second": 402.551,
+      "eval_steps_per_second": 50.379,
+      "step": 7000
+    },
+    {
+      "epoch": 0.33,
+      "learning_rate": 2.2211111111111114e-07,
+      "loss": 1.109,
+      "step": 8000
+    },
+    {
+      "epoch": 0.33,
+      "eval_accuracy": 0.3347305389221557,
+      "eval_loss": 1.098678469657898,
+      "eval_runtime": 12.5978,
+      "eval_samples_per_second": 397.688,
+      "eval_steps_per_second": 49.771,
+      "step": 8000
+    },
+    {
+      "epoch": 0.37,
+      "learning_rate": 2.4986111111111113e-07,
+      "loss": 1.1117,
+      "step": 9000
+    },
+    {
+      "epoch": 0.37,
+      "eval_accuracy": 0.34311377245508984,
+      "eval_loss": 1.0984419584274292,
+      "eval_runtime": 12.313,
+      "eval_samples_per_second": 406.886,
+      "eval_steps_per_second": 50.922,
+      "step": 9000
+    },
+    {
+      "epoch": 0.41,
+      "learning_rate": 2.776388888888889e-07,
+      "loss": 1.1081,
+      "step": 10000
+    },
+    {
+      "epoch": 0.41,
+      "eval_accuracy": 0.35429141716566864,
+      "eval_loss": 1.0968743562698364,
+      "eval_runtime": 12.4088,
+      "eval_samples_per_second": 403.744,
+      "eval_steps_per_second": 50.528,
+      "step": 10000
+    },
+    {
+      "epoch": 0.45,
+      "learning_rate": 3.053888888888889e-07,
+      "loss": 1.107,
+      "step": 11000
+    },
+    {
+      "epoch": 0.45,
+      "eval_accuracy": 0.3722554890219561,
+      "eval_loss": 1.0950665473937988,
+      "eval_runtime": 12.5775,
+      "eval_samples_per_second": 398.33,
+      "eval_steps_per_second": 49.851,
+      "step": 11000
+    },
+    {
+      "epoch": 0.49,
+      "learning_rate": 3.331666666666667e-07,
+      "loss": 1.1057,
+      "step": 12000
+    },
+    {
+      "epoch": 0.49,
+      "eval_accuracy": 0.37544910179640717,
+      "eval_loss": 1.0940738916397095,
+      "eval_runtime": 12.5186,
+      "eval_samples_per_second": 400.203,
+      "eval_steps_per_second": 50.085,
+      "step": 12000
+    },
+    {
+      "epoch": 0.53,
+      "learning_rate": 3.609166666666667e-07,
+      "loss": 1.1069,
+      "step": 13000
+    },
+    {
+      "epoch": 0.53,
+      "eval_accuracy": 0.3401197604790419,
+      "eval_loss": 1.0937162637710571,
+      "eval_runtime": 12.2594,
+      "eval_samples_per_second": 408.666,
+      "eval_steps_per_second": 51.144,
+      "step": 13000
+    },
+    {
+      "epoch": 0.57,
+      "learning_rate": 3.886944444444445e-07,
+      "loss": 1.1046,
+      "step": 14000
+    },
+    {
+      "epoch": 0.57,
+      "eval_accuracy": 0.37604790419161677,
+      "eval_loss": 1.0926916599273682,
+      "eval_runtime": 12.256,
+      "eval_samples_per_second": 408.78,
+      "eval_steps_per_second": 51.159,
+      "step": 14000
+    },
+    {
+      "epoch": 0.61,
+      "learning_rate": 4.164444444444445e-07,
+      "loss": 1.1046,
+      "step": 15000
+    },
+    {
+      "epoch": 0.61,
+      "eval_accuracy": 0.39201596806387223,
+      "eval_loss": 1.0912976264953613,
+      "eval_runtime": 12.5019,
+      "eval_samples_per_second": 400.74,
+      "eval_steps_per_second": 50.152,
+      "step": 15000
+    },
+    {
+      "epoch": 0.65,
+      "learning_rate": 4.442222222222223e-07,
+      "loss": 1.1052,
+      "step": 16000
+    },
+    {
+      "epoch": 0.65,
+      "eval_accuracy": 0.393812375249501,
+      "eval_loss": 1.0901482105255127,
+      "eval_runtime": 12.3407,
+      "eval_samples_per_second": 405.973,
+      "eval_steps_per_second": 50.807,
+      "step": 16000
+    },
+    {
+      "epoch": 0.69,
+      "learning_rate": 4.7197222222222224e-07,
+      "loss": 1.1011,
+      "step": 17000
+    },
+    {
+      "epoch": 0.69,
+      "eval_accuracy": 0.3932135728542914,
+      "eval_loss": 1.0901678800582886,
+      "eval_runtime": 12.3644,
+      "eval_samples_per_second": 405.197,
+      "eval_steps_per_second": 50.71,
+      "step": 17000
+    }
+  ],
+  "logging_steps": 1000,
+  "max_steps": 10000000,
+  "num_train_epochs": 408,
+  "save_steps": 1000,
+  "total_flos": 5923475030016000.0,
+  "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:f48fd322744a41332d86e79cb68054df0053a36e9279c46ff138f13ef1aab365
+size 4027