Training in progress, step 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55dcf2d90f757876dbb7c4b94eea3ec94f5166a71b3fbe4f0e962072d7107aec
+oid sha256:be3fbf39b8eb79e3df9922f6763aa922164fc565b8a550581ae148e0984a5a12
 size 1340618660

run-2/checkpoint-5500/config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "bert-large-uncased",
+  "architectures": [
+    "BertForMultipleChoice"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

run-2/checkpoint-5500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0cd65baa3989faed829a25b4c49615cc3af773a29305c9d14811cd5d49b17a0b
+size 1340618660

run-2/checkpoint-5500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1154ccd7c5054b3a56a259bf9c992ecfcc6d4848a90422ca820107a86d9b14bc
+size 2681472237

run-2/checkpoint-5500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:300dc1c3d7ab7e4437afb87a59de115242ea05773ae84e44b4263db51b96c059
+size 14244

run-2/checkpoint-5500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:560d7b96f3a1f6cf569e1ac49a2d24bfff0f09cbe5489fc2d9a895179ead3f0d
+size 1064

run-2/checkpoint-5500/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

run-2/checkpoint-5500/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

run-2/checkpoint-5500/trainer_state.json

@@ -0,0 +1,164 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 7.372654155495979,
+  "eval_steps": 500,
+  "global_step": 5500,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.6702412868632708,
+      "grad_norm": 17.923147201538086,
+      "learning_rate": 1.3001640168623715e-05,
+      "loss": 0.5514,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7683539986610413,
+      "eval_loss": 0.46452510356903076,
+      "eval_runtime": 8.9706,
+      "eval_samples_per_second": 332.531,
+      "eval_steps_per_second": 20.846,
+      "step": 746
+    },
+    {
+      "epoch": 1.3404825737265416,
+      "grad_norm": 19.32151222229004,
+      "learning_rate": 1.2067614294440976e-05,
+      "loss": 0.3969,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7643312215805054,
+      "eval_loss": 0.5671606063842773,
+      "eval_runtime": 9.0896,
+      "eval_samples_per_second": 328.177,
+      "eval_steps_per_second": 20.573,
+      "step": 1492
+    },
+    {
+      "epoch": 2.0107238605898123,
+      "grad_norm": 17.702476501464844,
+      "learning_rate": 1.1133588420258238e-05,
+      "loss": 0.2998,
+      "step": 1500
+    },
+    {
+      "epoch": 2.680965147453083,
+      "grad_norm": 6.516891956329346,
+      "learning_rate": 1.0199562546075501e-05,
+      "loss": 0.1319,
+      "step": 2000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7770700454711914,
+      "eval_loss": 0.7936307191848755,
+      "eval_runtime": 9.0031,
+      "eval_samples_per_second": 331.332,
+      "eval_steps_per_second": 20.771,
+      "step": 2238
+    },
+    {
+      "epoch": 3.351206434316354,
+      "grad_norm": 0.19944968819618225,
+      "learning_rate": 9.265536671892763e-06,
+      "loss": 0.0932,
+      "step": 2500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7750586867332458,
+      "eval_loss": 1.1848183870315552,
+      "eval_runtime": 9.064,
+      "eval_samples_per_second": 329.104,
+      "eval_steps_per_second": 20.631,
+      "step": 2984
+    },
+    {
+      "epoch": 4.021447721179625,
+      "grad_norm": 0.03240065276622772,
+      "learning_rate": 8.331510797710023e-06,
+      "loss": 0.0618,
+      "step": 3000
+    },
+    {
+      "epoch": 4.6916890080428955,
+      "grad_norm": 23.173229217529297,
+      "learning_rate": 7.397484923527286e-06,
+      "loss": 0.0314,
+      "step": 3500
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7707006335258484,
+      "eval_loss": 1.372557520866394,
+      "eval_runtime": 9.0505,
+      "eval_samples_per_second": 329.594,
+      "eval_steps_per_second": 20.662,
+      "step": 3730
+    },
+    {
+      "epoch": 5.361930294906166,
+      "grad_norm": 0.01009325310587883,
+      "learning_rate": 6.463459049344548e-06,
+      "loss": 0.0281,
+      "step": 4000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.777405321598053,
+      "eval_loss": 1.31714928150177,
+      "eval_runtime": 9.0862,
+      "eval_samples_per_second": 328.299,
+      "eval_steps_per_second": 20.581,
+      "step": 4476
+    },
+    {
+      "epoch": 6.032171581769437,
+      "grad_norm": 0.6762986779212952,
+      "learning_rate": 5.529433175161809e-06,
+      "loss": 0.0202,
+      "step": 4500
+    },
+    {
+      "epoch": 6.702412868632708,
+      "grad_norm": 0.610667884349823,
+      "learning_rate": 4.595407300979072e-06,
+      "loss": 0.0144,
+      "step": 5000
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7767348289489746,
+      "eval_loss": 1.4879707098007202,
+      "eval_runtime": 8.9951,
+      "eval_samples_per_second": 331.626,
+      "eval_steps_per_second": 20.789,
+      "step": 5222
+    },
+    {
+      "epoch": 7.372654155495979,
+      "grad_norm": 0.01202669832855463,
+      "learning_rate": 3.6613814267963338e-06,
+      "loss": 0.0084,
+      "step": 5500
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7460,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 1.3056507223800444e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.3935666042806453e-05,
+    "per_device_train_batch_size": 16
+  }
+}

run-2/checkpoint-5500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c68cfe34dae325096495265befbc04045f52f535badd4f779e9c272d3d0b0f80
+size 5048

run-2/checkpoint-6000/config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "bert-large-uncased",
+  "architectures": [
+    "BertForMultipleChoice"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

run-2/checkpoint-6000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:077129055d44c172ca6183ba0418bc1498b41ca2a9990e64a77f1ef2d88d33cb
+size 1340618660

run-2/checkpoint-6000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44540076373569cb070d8e72dcf2618fccfa379d06778985c552894e50bf4472
+size 2681472237

run-2/checkpoint-6000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:944c8125038577a50af7b702b16b5474f45a4ed64c6478b1129337816adaf2dc
+size 14244

run-2/checkpoint-6000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4efc00862b61248ce5bf9247cc87f45d9f031c91d3a0bac35166ce41295f97d
+size 1064

run-2/checkpoint-6000/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

run-2/checkpoint-6000/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

run-2/checkpoint-6000/trainer_state.json

@@ -0,0 +1,180 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.04289544235925,
+  "eval_steps": 500,
+  "global_step": 6000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.6702412868632708,
+      "grad_norm": 17.923147201538086,
+      "learning_rate": 1.3001640168623715e-05,
+      "loss": 0.5514,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7683539986610413,
+      "eval_loss": 0.46452510356903076,
+      "eval_runtime": 8.9706,
+      "eval_samples_per_second": 332.531,
+      "eval_steps_per_second": 20.846,
+      "step": 746
+    },
+    {
+      "epoch": 1.3404825737265416,
+      "grad_norm": 19.32151222229004,
+      "learning_rate": 1.2067614294440976e-05,
+      "loss": 0.3969,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7643312215805054,
+      "eval_loss": 0.5671606063842773,
+      "eval_runtime": 9.0896,
+      "eval_samples_per_second": 328.177,
+      "eval_steps_per_second": 20.573,
+      "step": 1492
+    },
+    {
+      "epoch": 2.0107238605898123,
+      "grad_norm": 17.702476501464844,
+      "learning_rate": 1.1133588420258238e-05,
+      "loss": 0.2998,
+      "step": 1500
+    },
+    {
+      "epoch": 2.680965147453083,
+      "grad_norm": 6.516891956329346,
+      "learning_rate": 1.0199562546075501e-05,
+      "loss": 0.1319,
+      "step": 2000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7770700454711914,
+      "eval_loss": 0.7936307191848755,
+      "eval_runtime": 9.0031,
+      "eval_samples_per_second": 331.332,
+      "eval_steps_per_second": 20.771,
+      "step": 2238
+    },
+    {
+      "epoch": 3.351206434316354,
+      "grad_norm": 0.19944968819618225,
+      "learning_rate": 9.265536671892763e-06,
+      "loss": 0.0932,
+      "step": 2500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7750586867332458,
+      "eval_loss": 1.1848183870315552,
+      "eval_runtime": 9.064,
+      "eval_samples_per_second": 329.104,
+      "eval_steps_per_second": 20.631,
+      "step": 2984
+    },
+    {
+      "epoch": 4.021447721179625,
+      "grad_norm": 0.03240065276622772,
+      "learning_rate": 8.331510797710023e-06,
+      "loss": 0.0618,
+      "step": 3000
+    },
+    {
+      "epoch": 4.6916890080428955,
+      "grad_norm": 23.173229217529297,
+      "learning_rate": 7.397484923527286e-06,
+      "loss": 0.0314,
+      "step": 3500
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7707006335258484,
+      "eval_loss": 1.372557520866394,
+      "eval_runtime": 9.0505,
+      "eval_samples_per_second": 329.594,
+      "eval_steps_per_second": 20.662,
+      "step": 3730
+    },
+    {
+      "epoch": 5.361930294906166,
+      "grad_norm": 0.01009325310587883,
+      "learning_rate": 6.463459049344548e-06,
+      "loss": 0.0281,
+      "step": 4000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.777405321598053,
+      "eval_loss": 1.31714928150177,
+      "eval_runtime": 9.0862,
+      "eval_samples_per_second": 328.299,
+      "eval_steps_per_second": 20.581,
+      "step": 4476
+    },
+    {
+      "epoch": 6.032171581769437,
+      "grad_norm": 0.6762986779212952,
+      "learning_rate": 5.529433175161809e-06,
+      "loss": 0.0202,
+      "step": 4500
+    },
+    {
+      "epoch": 6.702412868632708,
+      "grad_norm": 0.610667884349823,
+      "learning_rate": 4.595407300979072e-06,
+      "loss": 0.0144,
+      "step": 5000
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7767348289489746,
+      "eval_loss": 1.4879707098007202,
+      "eval_runtime": 8.9951,
+      "eval_samples_per_second": 331.626,
+      "eval_steps_per_second": 20.789,
+      "step": 5222
+    },
+    {
+      "epoch": 7.372654155495979,
+      "grad_norm": 0.01202669832855463,
+      "learning_rate": 3.6613814267963338e-06,
+      "loss": 0.0084,
+      "step": 5500
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7763996124267578,
+      "eval_loss": 1.6179240942001343,
+      "eval_runtime": 9.0802,
+      "eval_samples_per_second": 328.518,
+      "eval_steps_per_second": 20.594,
+      "step": 5968
+    },
+    {
+      "epoch": 8.04289544235925,
+      "grad_norm": 0.007873074151575565,
+      "learning_rate": 2.7273555526135954e-06,
+      "loss": 0.0046,
+      "step": 6000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7460,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 1.4239986889703832e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.3935666042806453e-05,
+    "per_device_train_batch_size": 16
+  }
+}

run-2/checkpoint-6000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c68cfe34dae325096495265befbc04045f52f535badd4f779e9c272d3d0b0f80
+size 5048

run-2/checkpoint-6500/config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "bert-large-uncased",
+  "architectures": [
+    "BertForMultipleChoice"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

run-2/checkpoint-6500/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06d9c4a077c9d50157ebfc56ce852e14881aeba11d715b15c06afecdb067dbf1
+size 1340618660

run-2/checkpoint-6500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1d888f854d88a70b3fdf31db01aa04f5c53b9fb4b6974f4e7b5629fa789ad93
+size 2681472237

run-2/checkpoint-6500/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7d2f0eef300c725d58e5b583509e78cdc2ebab68ccb82336ef58b37c3ad6272
+size 14244

run-2/checkpoint-6500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf21db80d2360643f33f05d3f9c9bce21878acef71716412589e941a030859ec
+size 1064

run-2/checkpoint-6500/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

run-2/checkpoint-6500/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

run-2/checkpoint-6500/trainer_state.json

@@ -0,0 +1,187 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.71313672922252,
+  "eval_steps": 500,
+  "global_step": 6500,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.6702412868632708,
+      "grad_norm": 17.923147201538086,
+      "learning_rate": 1.3001640168623715e-05,
+      "loss": 0.5514,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7683539986610413,
+      "eval_loss": 0.46452510356903076,
+      "eval_runtime": 8.9706,
+      "eval_samples_per_second": 332.531,
+      "eval_steps_per_second": 20.846,
+      "step": 746
+    },
+    {
+      "epoch": 1.3404825737265416,
+      "grad_norm": 19.32151222229004,
+      "learning_rate": 1.2067614294440976e-05,
+      "loss": 0.3969,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7643312215805054,
+      "eval_loss": 0.5671606063842773,
+      "eval_runtime": 9.0896,
+      "eval_samples_per_second": 328.177,
+      "eval_steps_per_second": 20.573,
+      "step": 1492
+    },
+    {
+      "epoch": 2.0107238605898123,
+      "grad_norm": 17.702476501464844,
+      "learning_rate": 1.1133588420258238e-05,
+      "loss": 0.2998,
+      "step": 1500
+    },
+    {
+      "epoch": 2.680965147453083,
+      "grad_norm": 6.516891956329346,
+      "learning_rate": 1.0199562546075501e-05,
+      "loss": 0.1319,
+      "step": 2000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7770700454711914,
+      "eval_loss": 0.7936307191848755,
+      "eval_runtime": 9.0031,
+      "eval_samples_per_second": 331.332,
+      "eval_steps_per_second": 20.771,
+      "step": 2238
+    },
+    {
+      "epoch": 3.351206434316354,
+      "grad_norm": 0.19944968819618225,
+      "learning_rate": 9.265536671892763e-06,
+      "loss": 0.0932,
+      "step": 2500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7750586867332458,
+      "eval_loss": 1.1848183870315552,
+      "eval_runtime": 9.064,
+      "eval_samples_per_second": 329.104,
+      "eval_steps_per_second": 20.631,
+      "step": 2984
+    },
+    {
+      "epoch": 4.021447721179625,
+      "grad_norm": 0.03240065276622772,
+      "learning_rate": 8.331510797710023e-06,
+      "loss": 0.0618,
+      "step": 3000
+    },
+    {
+      "epoch": 4.6916890080428955,
+      "grad_norm": 23.173229217529297,
+      "learning_rate": 7.397484923527286e-06,
+      "loss": 0.0314,
+      "step": 3500
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7707006335258484,
+      "eval_loss": 1.372557520866394,
+      "eval_runtime": 9.0505,
+      "eval_samples_per_second": 329.594,
+      "eval_steps_per_second": 20.662,
+      "step": 3730
+    },
+    {
+      "epoch": 5.361930294906166,
+      "grad_norm": 0.01009325310587883,
+      "learning_rate": 6.463459049344548e-06,
+      "loss": 0.0281,
+      "step": 4000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.777405321598053,
+      "eval_loss": 1.31714928150177,
+      "eval_runtime": 9.0862,
+      "eval_samples_per_second": 328.299,
+      "eval_steps_per_second": 20.581,
+      "step": 4476
+    },
+    {
+      "epoch": 6.032171581769437,
+      "grad_norm": 0.6762986779212952,
+      "learning_rate": 5.529433175161809e-06,
+      "loss": 0.0202,
+      "step": 4500
+    },
+    {
+      "epoch": 6.702412868632708,
+      "grad_norm": 0.610667884349823,
+      "learning_rate": 4.595407300979072e-06,
+      "loss": 0.0144,
+      "step": 5000
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7767348289489746,
+      "eval_loss": 1.4879707098007202,
+      "eval_runtime": 8.9951,
+      "eval_samples_per_second": 331.626,
+      "eval_steps_per_second": 20.789,
+      "step": 5222
+    },
+    {
+      "epoch": 7.372654155495979,
+      "grad_norm": 0.01202669832855463,
+      "learning_rate": 3.6613814267963338e-06,
+      "loss": 0.0084,
+      "step": 5500
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7763996124267578,
+      "eval_loss": 1.6179240942001343,
+      "eval_runtime": 9.0802,
+      "eval_samples_per_second": 328.518,
+      "eval_steps_per_second": 20.594,
+      "step": 5968
+    },
+    {
+      "epoch": 8.04289544235925,
+      "grad_norm": 0.007873074151575565,
+      "learning_rate": 2.7273555526135954e-06,
+      "loss": 0.0046,
+      "step": 6000
+    },
+    {
+      "epoch": 8.71313672922252,
+      "grad_norm": 0.029083162546157837,
+      "learning_rate": 1.7933296784308572e-06,
+      "loss": 0.0044,
+      "step": 6500
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7460,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 1.5429010800414744e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.3935666042806453e-05,
+    "per_device_train_batch_size": 16
+  }
+}

run-2/checkpoint-6500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c68cfe34dae325096495265befbc04045f52f535badd4f779e9c272d3d0b0f80
+size 5048

run-2/checkpoint-7000/config.json

@@ -0,0 +1,26 @@
+{
+  "_name_or_path": "bert-large-uncased",
+  "architectures": [
+    "BertForMultipleChoice"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "classifier_dropout": null,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layer_norm_eps": 1e-12,
+  "max_position_embeddings": 512,
+  "model_type": "bert",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "pad_token_id": 0,
+  "position_embedding_type": "absolute",
+  "torch_dtype": "float32",
+  "transformers_version": "4.40.1",
+  "type_vocab_size": 2,
+  "use_cache": true,
+  "vocab_size": 30522
+}

run-2/checkpoint-7000/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1aec656b5dc6ccada7db9c4d9734b84467f68badbcfc003fe48de4a7ed40732f
+size 1340618660

run-2/checkpoint-7000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efc90298d9e8b897490d54fa9b155708d7b889e24518597b184aa65ba8f006f6
+size 2681472237

run-2/checkpoint-7000/rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6671eaad5d0f1524db8436098c89faae10d33bc9f9fe6f85d5cc287d3530603d
+size 14244

run-2/checkpoint-7000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35eb178d41178ab3f7b40c27735dfe3e38092325602f092b3009b0950df836a8
+size 1064

run-2/checkpoint-7000/special_tokens_map.json

@@ -0,0 +1,7 @@
+{
+  "cls_token": "[CLS]",
+  "mask_token": "[MASK]",
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "unk_token": "[UNK]"
+}

run-2/checkpoint-7000/tokenizer_config.json

@@ -0,0 +1,55 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "100": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "101": {
+      "content": "[CLS]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "102": {
+      "content": "[SEP]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "103": {
+      "content": "[MASK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "do_lower_case": true,
+  "mask_token": "[MASK]",
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "strip_accents": null,
+  "tokenize_chinese_chars": true,
+  "tokenizer_class": "BertTokenizer",
+  "unk_token": "[UNK]"
+}

run-2/checkpoint-7000/trainer_state.json

@@ -0,0 +1,203 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 9.383378016085791,
+  "eval_steps": 500,
+  "global_step": 7000,
+  "is_hyper_param_search": true,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.6702412868632708,
+      "grad_norm": 17.923147201538086,
+      "learning_rate": 1.3001640168623715e-05,
+      "loss": 0.5514,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.7683539986610413,
+      "eval_loss": 0.46452510356903076,
+      "eval_runtime": 8.9706,
+      "eval_samples_per_second": 332.531,
+      "eval_steps_per_second": 20.846,
+      "step": 746
+    },
+    {
+      "epoch": 1.3404825737265416,
+      "grad_norm": 19.32151222229004,
+      "learning_rate": 1.2067614294440976e-05,
+      "loss": 0.3969,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.7643312215805054,
+      "eval_loss": 0.5671606063842773,
+      "eval_runtime": 9.0896,
+      "eval_samples_per_second": 328.177,
+      "eval_steps_per_second": 20.573,
+      "step": 1492
+    },
+    {
+      "epoch": 2.0107238605898123,
+      "grad_norm": 17.702476501464844,
+      "learning_rate": 1.1133588420258238e-05,
+      "loss": 0.2998,
+      "step": 1500
+    },
+    {
+      "epoch": 2.680965147453083,
+      "grad_norm": 6.516891956329346,
+      "learning_rate": 1.0199562546075501e-05,
+      "loss": 0.1319,
+      "step": 2000
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.7770700454711914,
+      "eval_loss": 0.7936307191848755,
+      "eval_runtime": 9.0031,
+      "eval_samples_per_second": 331.332,
+      "eval_steps_per_second": 20.771,
+      "step": 2238
+    },
+    {
+      "epoch": 3.351206434316354,
+      "grad_norm": 0.19944968819618225,
+      "learning_rate": 9.265536671892763e-06,
+      "loss": 0.0932,
+      "step": 2500
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.7750586867332458,
+      "eval_loss": 1.1848183870315552,
+      "eval_runtime": 9.064,
+      "eval_samples_per_second": 329.104,
+      "eval_steps_per_second": 20.631,
+      "step": 2984
+    },
+    {
+      "epoch": 4.021447721179625,
+      "grad_norm": 0.03240065276622772,
+      "learning_rate": 8.331510797710023e-06,
+      "loss": 0.0618,
+      "step": 3000
+    },
+    {
+      "epoch": 4.6916890080428955,
+      "grad_norm": 23.173229217529297,
+      "learning_rate": 7.397484923527286e-06,
+      "loss": 0.0314,
+      "step": 3500
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.7707006335258484,
+      "eval_loss": 1.372557520866394,
+      "eval_runtime": 9.0505,
+      "eval_samples_per_second": 329.594,
+      "eval_steps_per_second": 20.662,
+      "step": 3730
+    },
+    {
+      "epoch": 5.361930294906166,
+      "grad_norm": 0.01009325310587883,
+      "learning_rate": 6.463459049344548e-06,
+      "loss": 0.0281,
+      "step": 4000
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.777405321598053,
+      "eval_loss": 1.31714928150177,
+      "eval_runtime": 9.0862,
+      "eval_samples_per_second": 328.299,
+      "eval_steps_per_second": 20.581,
+      "step": 4476
+    },
+    {
+      "epoch": 6.032171581769437,
+      "grad_norm": 0.6762986779212952,
+      "learning_rate": 5.529433175161809e-06,
+      "loss": 0.0202,
+      "step": 4500
+    },
+    {
+      "epoch": 6.702412868632708,
+      "grad_norm": 0.610667884349823,
+      "learning_rate": 4.595407300979072e-06,
+      "loss": 0.0144,
+      "step": 5000
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.7767348289489746,
+      "eval_loss": 1.4879707098007202,
+      "eval_runtime": 8.9951,
+      "eval_samples_per_second": 331.626,
+      "eval_steps_per_second": 20.789,
+      "step": 5222
+    },
+    {
+      "epoch": 7.372654155495979,
+      "grad_norm": 0.01202669832855463,
+      "learning_rate": 3.6613814267963338e-06,
+      "loss": 0.0084,
+      "step": 5500
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.7763996124267578,
+      "eval_loss": 1.6179240942001343,
+      "eval_runtime": 9.0802,
+      "eval_samples_per_second": 328.518,
+      "eval_steps_per_second": 20.594,
+      "step": 5968
+    },
+    {
+      "epoch": 8.04289544235925,
+      "grad_norm": 0.007873074151575565,
+      "learning_rate": 2.7273555526135954e-06,
+      "loss": 0.0046,
+      "step": 6000
+    },
+    {
+      "epoch": 8.71313672922252,
+      "grad_norm": 0.029083162546157837,
+      "learning_rate": 1.7933296784308572e-06,
+      "loss": 0.0044,
+      "step": 6500
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.777405321598053,
+      "eval_loss": 1.6098750829696655,
+      "eval_runtime": 9.0257,
+      "eval_samples_per_second": 330.499,
+      "eval_steps_per_second": 20.719,
+      "step": 6714
+    },
+    {
+      "epoch": 9.383378016085791,
+      "grad_norm": 0.14630918204784393,
+      "learning_rate": 8.59303804248119e-07,
+      "loss": 0.0057,
+      "step": 7000
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 7460,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 1.6607423106651636e+16,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": {
+    "learning_rate": 1.3935666042806453e-05,
+    "per_device_train_batch_size": 16
+  }
+}

run-2/checkpoint-7000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c68cfe34dae325096495265befbc04045f52f535badd4f779e9c272d3d0b0f80
+size 5048

run-3/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4329ff70638ffab5ce06e6a7170544445cd28f7109d1490e1191aa67cad82484
+oid sha256:4d9649ba9e265d8e4fe5d708f3e7adb9d6191f9844715240ec35d509ec8e5fe6
 size 1340618660

run-3/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:552d76adc3ad8e3fee3a8f35e7fac6e259f9c9347879331b6fdbf60c33d736b9
+oid sha256:b6ef50bc2eb08bb3dbb1e2f70d0fdf07d835c95837a40f398f4436b48b906838
 size 2681472237

run-3/checkpoint-1000/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f18c5d8cacab124798e04b8f21f29ee72f21611ea7bee24e198243f27a9b7b21
+oid sha256:f15ef18f77678b087c8d8d016723a03f0d0f4e837051c67af83ea5404cfb3e09
 size 14244

run-3/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fb220e76bba57e6a943673aae70413c1a81e7d6a5205dae366454ee0f4601c56
+oid sha256:f2e778733aea3d76f79d700f7557d56cc34f9c1df8dc23634cecfda952769b78
 size 1064

run-3/checkpoint-1000/trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.3404825737265416,
+  "epoch": 2.680965147453083,
   "eval_steps": 500,
   "global_step": 1000,
   "is_hyper_param_search": true,
@@ -9,39 +9,48 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.6702412868632708,
-      "grad_norm": 13.834343910217285,
-      "learning_rate": 1.8689758651553552e-05,
-      "loss": 0.5491,
+      "epoch": 1.0,
+      "eval_accuracy": 0.7727120518684387,
+      "eval_loss": 0.46057018637657166,
+      "eval_runtime": 8.9113,
+      "eval_samples_per_second": 334.743,
+      "eval_steps_per_second": 20.985,
+      "step": 373
+    },
+    {
+      "epoch": 1.3404825737265416,
+      "grad_norm": 8.923430442810059,
+      "learning_rate": 2.587666867882196e-05,
+      "loss": 0.4651,
       "step": 500
     },
     {
-      "epoch": 1.0,
-      "eval_accuracy": 0.762319803237915,
-      "eval_loss": 0.4773792326450348,
-      "eval_runtime": 8.5342,
-      "eval_samples_per_second": 349.533,
-      "eval_steps_per_second": 21.912,
+      "epoch": 2.0,
+      "eval_accuracy": 0.7770700454711914,
+      "eval_loss": 0.5375419855117798,
+      "eval_runtime": 9.0162,
+      "eval_samples_per_second": 330.848,
+      "eval_steps_per_second": 20.74,
       "step": 746
     },
     {
-      "epoch": 1.3404825737265416,
-      "grad_norm": 12.70506477355957,
-      "learning_rate": 1.7347103576010912e-05,
-      "loss": 0.4116,
+      "epoch": 2.680965147453083,
+      "grad_norm": 30.215883255004883,
+      "learning_rate": 2.187099241275045e-05,
+      "loss": 0.176,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 7460,
+  "max_steps": 3730,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 10,
   "save_steps": 500,
-  "total_flos": 2363719425123456.0,
-  "train_batch_size": 16,
+  "total_flos": 5069168917756236.0,
+  "train_batch_size": 32,
   "trial_name": null,
   "trial_params": {
-    "learning_rate": 2.0032413727096193e-05,
-    "per_device_train_batch_size": 16
+    "learning_rate": 2.988234494489347e-05,
+    "per_device_train_batch_size": 32
   }
 }