[Model] roberta-base-cola

config.json

@@ -0,0 +1,37 @@
+{
+  "_name_or_path": "/mnt/lustre/zhangyunchen/transformers/roberta-base",
+  "architectures": [
+    "RobertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "finetuning_task": "cola",
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "id2label": {
+    "0": "unacceptable",
+    "1": "acceptable"
+  },
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "label2id": {
+    "acceptable": 1,
+    "unacceptable": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "torch_dtype": "float32",
+  "transformers_version": "4.16.2",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ce93ed6b663738f27839273e760adec72615c032f231c691737639c0810f69c
+size 997304355

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:430376113b7f419456191f13afe5f30b1e05712113747f382aa9650abf108348
+size 498679699

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36cd3ada4a2289ccec5b1a52b18da87e3668212e174d14fbb8e3312debf2dcef
+size 14659

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:33e2d14f09dacec4f4ffd0850b4478e0e48fd13dc3bd297f3b0ee1087e310f7e
+size 623

special_tokens_map.json

@@ -0,0 +1 @@
+{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": {"content": "<mask>", "single_word": false, "lstrip": true, "rstrip": false, "normalized": false}}

tokenizer_config.json

@@ -0,0 +1 @@
+{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "trim_offsets": true, "special_tokens_map_file": null, "name_or_path": "/mnt/lustre/zhangyunchen/transformers/roberta-base", "tokenizer_class": "RobertaTokenizer"}

trainer_state.json

@@ -0,0 +1,211 @@
+{
+  "best_metric": 0.6249925495853809,
+  "best_model_checkpoint": "./fp32_2e_5/models/cola-roberta-base/checkpoint-1300",
+  "epoch": 4.850746268656716,
+  "global_step": 1300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.37,
+      "learning_rate": 1.2422360248447205e-05,
+      "loss": 0.6277,
+      "step": 100
+    },
+    {
+      "epoch": 0.37,
+      "eval_loss": 0.5836151242256165,
+      "eval_matthews_correlation": 0.0,
+      "eval_runtime": 4.0206,
+      "eval_samples_per_second": 259.413,
+      "eval_steps_per_second": 8.208,
+      "step": 100
+    },
+    {
+      "epoch": 0.75,
+      "learning_rate": 1.9690353314807464e-05,
+      "loss": 0.508,
+      "step": 200
+    },
+    {
+      "epoch": 0.75,
+      "eval_loss": 0.4776883125305176,
+      "eval_matthews_correlation": 0.49587290918084903,
+      "eval_runtime": 2.8206,
+      "eval_samples_per_second": 369.773,
+      "eval_steps_per_second": 11.699,
+      "step": 200
+    },
+    {
+      "epoch": 1.12,
+      "learning_rate": 1.8896387455339422e-05,
+      "loss": 0.4118,
+      "step": 300
+    },
+    {
+      "epoch": 1.12,
+      "eval_loss": 0.430734246969223,
+      "eval_matthews_correlation": 0.5308757570358055,
+      "eval_runtime": 2.756,
+      "eval_samples_per_second": 378.442,
+      "eval_steps_per_second": 11.974,
+      "step": 300
+    },
+    {
+      "epoch": 1.49,
+      "learning_rate": 1.810242159587138e-05,
+      "loss": 0.3547,
+      "step": 400
+    },
+    {
+      "epoch": 1.49,
+      "eval_loss": 0.5029284954071045,
+      "eval_matthews_correlation": 0.5190360820585255,
+      "eval_runtime": 2.8845,
+      "eval_samples_per_second": 361.594,
+      "eval_steps_per_second": 11.441,
+      "step": 400
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.7308455736403335e-05,
+      "loss": 0.3569,
+      "step": 500
+    },
+    {
+      "epoch": 1.87,
+      "eval_loss": 0.3902229964733124,
+      "eval_matthews_correlation": 0.5946035554045278,
+      "eval_runtime": 2.7094,
+      "eval_samples_per_second": 384.951,
+      "eval_steps_per_second": 12.18,
+      "step": 500
+    },
+    {
+      "epoch": 2.24,
+      "learning_rate": 1.6514489876935293e-05,
+      "loss": 0.2896,
+      "step": 600
+    },
+    {
+      "epoch": 2.24,
+      "eval_loss": 0.49517160654067993,
+      "eval_matthews_correlation": 0.5881955703622721,
+      "eval_runtime": 2.742,
+      "eval_samples_per_second": 380.386,
+      "eval_steps_per_second": 12.035,
+      "step": 600
+    },
+    {
+      "epoch": 2.61,
+      "learning_rate": 1.5720524017467248e-05,
+      "loss": 0.2368,
+      "step": 700
+    },
+    {
+      "epoch": 2.61,
+      "eval_loss": 0.5459012985229492,
+      "eval_matthews_correlation": 0.567550266689718,
+      "eval_runtime": 2.7779,
+      "eval_samples_per_second": 375.462,
+      "eval_steps_per_second": 11.879,
+      "step": 700
+    },
+    {
+      "epoch": 2.99,
+      "learning_rate": 1.4926558157999208e-05,
+      "loss": 0.2583,
+      "step": 800
+    },
+    {
+      "epoch": 2.99,
+      "eval_loss": 0.531144380569458,
+      "eval_matthews_correlation": 0.5653000167449913,
+      "eval_runtime": 2.8725,
+      "eval_samples_per_second": 363.101,
+      "eval_steps_per_second": 11.488,
+      "step": 800
+    },
+    {
+      "epoch": 3.36,
+      "learning_rate": 1.4132592298531164e-05,
+      "loss": 0.1631,
+      "step": 900
+    },
+    {
+      "epoch": 3.36,
+      "eval_loss": 0.8727195262908936,
+      "eval_matthews_correlation": 0.5115610675474672,
+      "eval_runtime": 2.6938,
+      "eval_samples_per_second": 387.184,
+      "eval_steps_per_second": 12.25,
+      "step": 900
+    },
+    {
+      "epoch": 3.73,
+      "learning_rate": 1.333862643906312e-05,
+      "loss": 0.1758,
+      "step": 1000
+    },
+    {
+      "epoch": 3.73,
+      "eval_loss": 0.4821915030479431,
+      "eval_matthews_correlation": 0.6070286836001448,
+      "eval_runtime": 2.7002,
+      "eval_samples_per_second": 386.266,
+      "eval_steps_per_second": 12.221,
+      "step": 1000
+    },
+    {
+      "epoch": 4.1,
+      "learning_rate": 1.2544660579595079e-05,
+      "loss": 0.1694,
+      "step": 1100
+    },
+    {
+      "epoch": 4.1,
+      "eval_loss": 0.654831051826477,
+      "eval_matthews_correlation": 0.6021802636668068,
+      "eval_runtime": 2.7489,
+      "eval_samples_per_second": 379.425,
+      "eval_steps_per_second": 12.005,
+      "step": 1100
+    },
+    {
+      "epoch": 4.48,
+      "learning_rate": 1.1750694720127037e-05,
+      "loss": 0.135,
+      "step": 1200
+    },
+    {
+      "epoch": 4.48,
+      "eval_loss": 0.7147157788276672,
+      "eval_matthews_correlation": 0.5803609526755046,
+      "eval_runtime": 2.9189,
+      "eval_samples_per_second": 357.331,
+      "eval_steps_per_second": 11.306,
+      "step": 1200
+    },
+    {
+      "epoch": 4.85,
+      "learning_rate": 1.0956728860658991e-05,
+      "loss": 0.1459,
+      "step": 1300
+    },
+    {
+      "epoch": 4.85,
+      "eval_loss": 0.565829336643219,
+      "eval_matthews_correlation": 0.6249925495853809,
+      "eval_runtime": 2.7785,
+      "eval_samples_per_second": 375.382,
+      "eval_steps_per_second": 11.877,
+      "step": 1300
+    }
+  ],
+  "max_steps": 2680,
+  "num_train_epochs": 10,
+  "total_flos": 2729777199360000.0,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6a9ec72ff30398a3a57fb04827f755f976ef3eeef2f302dbe4134bc1a0ae726
+size 3055