Training in progress, epoch 3, checkpoint

checkpoint-300/config.json

@@ -0,0 +1,33 @@
+{
+  "_name_or_path": "microsoft/deberta-base",
+  "architectures": [
+    "DebertaForSequenceClassification"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "layer_norm_eps": 1e-07,
+  "legacy": true,
+  "max_position_embeddings": 512,
+  "max_relative_positions": -1,
+  "model_type": "deberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 0,
+  "pooler_dropout": 0,
+  "pooler_hidden_act": "gelu",
+  "pooler_hidden_size": 768,
+  "pos_att_type": [
+    "c2p",
+    "p2c"
+  ],
+  "position_biased_input": false,
+  "relative_attention": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.48.3",
+  "type_vocab_size": 0,
+  "vocab_size": 50265
+}

checkpoint-300/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:074dc5a628cb04cd83d08121e95f5cb66b1574ccdae323556dac31fabaab0cb7
+size 556799560

checkpoint-300/trainer_state.json

@@ -0,0 +1,171 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "eval_steps": 50,
+  "global_step": 300,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5,
+      "grad_norm": 5.4627485275268555,
+      "learning_rate": 2.75e-05,
+      "loss": 0.4747,
+      "step": 50
+    },
+    {
+      "epoch": 0.5,
+      "eval_pet_AUC": 0.9049348732937939,
+      "eval_pet_accuracy": 0.835,
+      "eval_pet_f1": 0.7053571428571429,
+      "eval_pet_loss": 0.41319525241851807,
+      "eval_pet_macro_f1": 0.7953869047619048,
+      "eval_pet_micro_f1": 0.835,
+      "eval_pet_precision": 0.8494623655913979,
+      "eval_pet_recall": 0.6030534351145038,
+      "eval_pet_runtime": 0.4406,
+      "eval_pet_samples_per_second": 907.941,
+      "eval_pet_specificity": 0.9479553903345725,
+      "eval_pet_steps_per_second": 2.27,
+      "step": 50
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 8.922032356262207,
+      "learning_rate": 2.5e-05,
+      "loss": 0.3861,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "eval_pet_AUC": 0.9257356905701071,
+      "eval_pet_accuracy": 0.8375,
+      "eval_pet_f1": 0.7735191637630662,
+      "eval_pet_loss": 0.3712470531463623,
+      "eval_pet_macro_f1": 0.8234067553708118,
+      "eval_pet_micro_f1": 0.8375,
+      "eval_pet_precision": 0.7115384615384616,
+      "eval_pet_recall": 0.8473282442748091,
+      "eval_pet_runtime": 0.4548,
+      "eval_pet_samples_per_second": 879.594,
+      "eval_pet_specificity": 0.8327137546468402,
+      "eval_pet_steps_per_second": 2.199,
+      "step": 100
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 11.44329833984375,
+      "learning_rate": 2.25e-05,
+      "loss": 0.2569,
+      "step": 150
+    },
+    {
+      "epoch": 1.5,
+      "eval_pet_AUC": 0.9245722069298221,
+      "eval_pet_accuracy": 0.8275,
+      "eval_pet_f1": 0.770764119601329,
+      "eval_pet_loss": 0.459354430437088,
+      "eval_pet_macro_f1": 0.8162437832475582,
+      "eval_pet_micro_f1": 0.8275,
+      "eval_pet_precision": 0.6823529411764706,
+      "eval_pet_recall": 0.8854961832061069,
+      "eval_pet_runtime": 0.4438,
+      "eval_pet_samples_per_second": 901.388,
+      "eval_pet_specificity": 0.7992565055762082,
+      "eval_pet_steps_per_second": 2.253,
+      "step": 150
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 4.656713962554932,
+      "learning_rate": 1.9999999999999998e-05,
+      "loss": 0.2611,
+      "step": 200
+    },
+    {
+      "epoch": 2.0,
+      "eval_pet_AUC": 0.9204858253639434,
+      "eval_pet_accuracy": 0.8425,
+      "eval_pet_f1": 0.7364016736401674,
+      "eval_pet_loss": 0.36228087544441223,
+      "eval_pet_macro_f1": 0.8120511041997629,
+      "eval_pet_micro_f1": 0.8425,
+      "eval_pet_precision": 0.8148148148148148,
+      "eval_pet_recall": 0.6717557251908397,
+      "eval_pet_runtime": 0.4544,
+      "eval_pet_samples_per_second": 880.368,
+      "eval_pet_specificity": 0.9256505576208178,
+      "eval_pet_steps_per_second": 2.201,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 14.081766128540039,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 0.185,
+      "step": 250
+    },
+    {
+      "epoch": 2.5,
+      "eval_pet_AUC": 0.927551860154942,
+      "eval_pet_accuracy": 0.8625,
+      "eval_pet_f1": 0.8042704626334519,
+      "eval_pet_loss": 0.45289406180381775,
+      "eval_pet_macro_f1": 0.8491487187926412,
+      "eval_pet_micro_f1": 0.8625,
+      "eval_pet_precision": 0.7533333333333333,
+      "eval_pet_recall": 0.8625954198473282,
+      "eval_pet_runtime": 0.4443,
+      "eval_pet_samples_per_second": 900.384,
+      "eval_pet_specificity": 0.862453531598513,
+      "eval_pet_steps_per_second": 2.251,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.8663891553878784,
+      "learning_rate": 1.5e-05,
+      "loss": 0.116,
+      "step": 300
+    },
+    {
+      "epoch": 3.0,
+      "eval_pet_AUC": 0.9217060643037545,
+      "eval_pet_accuracy": 0.85,
+      "eval_pet_f1": 0.7777777777777778,
+      "eval_pet_loss": 0.6567688584327698,
+      "eval_pet_macro_f1": 0.8322851153039832,
+      "eval_pet_micro_f1": 0.85,
+      "eval_pet_precision": 0.7553956834532374,
+      "eval_pet_recall": 0.8015267175572519,
+      "eval_pet_runtime": 0.4278,
+      "eval_pet_samples_per_second": 934.956,
+      "eval_pet_specificity": 0.8736059479553904,
+      "eval_pet_steps_per_second": 2.337,
+      "step": 300
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 600,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 114974099712000.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-300/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1eff87705db47f60992b6edf2d9be2f07821834faa22357994717fcc6d623de0
+size 5432