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 sha256:75edba797b0fff99b1a6a0977c6ec968dce21dd884871dd3e2385c737b3245ee
+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": 4.128329277038574,
+      "learning_rate": 9.166666666666666e-06,
+      "loss": 0.5694,
+      "step": 50
+    },
+    {
+      "epoch": 0.5,
+      "eval_pet_AUC": 0.877852385631619,
+      "eval_pet_accuracy": 0.7775,
+      "eval_pet_f1": 0.5898617511520737,
+      "eval_pet_loss": 0.4617109000682831,
+      "eval_pet_macro_f1": 0.7186015445297247,
+      "eval_pet_micro_f1": 0.7775,
+      "eval_pet_precision": 0.8533333333333334,
+      "eval_pet_recall": 0.4507042253521127,
+      "eval_pet_runtime": 0.434,
+      "eval_pet_samples_per_second": 921.733,
+      "eval_pet_specificity": 0.9573643410852714,
+      "eval_pet_steps_per_second": 2.304,
+      "step": 50
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 6.654369354248047,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.4313,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "eval_pet_AUC": 0.9084779997816355,
+      "eval_pet_accuracy": 0.815,
+      "eval_pet_f1": 0.7086614173228346,
+      "eval_pet_loss": 0.3827367424964905,
+      "eval_pet_macro_f1": 0.7865651408958496,
+      "eval_pet_micro_f1": 0.815,
+      "eval_pet_precision": 0.8035714285714286,
+      "eval_pet_recall": 0.6338028169014085,
+      "eval_pet_runtime": 0.4369,
+      "eval_pet_samples_per_second": 915.512,
+      "eval_pet_specificity": 0.9147286821705426,
+      "eval_pet_steps_per_second": 2.289,
+      "step": 100
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 3.3484041690826416,
+      "learning_rate": 7.500000000000001e-06,
+      "loss": 0.3259,
+      "step": 150
+    },
+    {
+      "epoch": 1.5,
+      "eval_pet_AUC": 0.9165028933289661,
+      "eval_pet_accuracy": 0.8325,
+      "eval_pet_f1": 0.7681660899653979,
+      "eval_pet_loss": 0.36348748207092285,
+      "eval_pet_macro_f1": 0.8185253150414074,
+      "eval_pet_micro_f1": 0.8325,
+      "eval_pet_precision": 0.7551020408163265,
+      "eval_pet_recall": 0.7816901408450704,
+      "eval_pet_runtime": 0.4775,
+      "eval_pet_samples_per_second": 837.78,
+      "eval_pet_specificity": 0.8604651162790697,
+      "eval_pet_steps_per_second": 2.094,
+      "step": 150
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 5.001344203948975,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.2924,
+      "step": 200
+    },
+    {
+      "epoch": 2.0,
+      "eval_pet_AUC": 0.9257287913527678,
+      "eval_pet_accuracy": 0.85,
+      "eval_pet_f1": 0.8,
+      "eval_pet_loss": 0.346834272146225,
+      "eval_pet_macro_f1": 0.8400000000000001,
+      "eval_pet_micro_f1": 0.85,
+      "eval_pet_precision": 0.759493670886076,
+      "eval_pet_recall": 0.8450704225352113,
+      "eval_pet_runtime": 0.4621,
+      "eval_pet_samples_per_second": 865.539,
+      "eval_pet_specificity": 0.8527131782945736,
+      "eval_pet_steps_per_second": 2.164,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 9.390085220336914,
+      "learning_rate": 5.833333333333334e-06,
+      "loss": 0.2524,
+      "step": 250
+    },
+    {
+      "epoch": 2.5,
+      "eval_pet_AUC": 0.9255377224587837,
+      "eval_pet_accuracy": 0.8275,
+      "eval_pet_f1": 0.7206477732793523,
+      "eval_pet_loss": 0.40142711997032166,
+      "eval_pet_macro_f1": 0.797936906531177,
+      "eval_pet_micro_f1": 0.8275,
+      "eval_pet_precision": 0.8476190476190476,
+      "eval_pet_recall": 0.6267605633802817,
+      "eval_pet_runtime": 0.4565,
+      "eval_pet_samples_per_second": 876.237,
+      "eval_pet_specificity": 0.937984496124031,
+      "eval_pet_steps_per_second": 2.191,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 5.205839157104492,
+      "learning_rate": 5e-06,
+      "loss": 0.2088,
+      "step": 300
+    },
+    {
+      "epoch": 3.0,
+      "eval_pet_AUC": 0.9287313025439459,
+      "eval_pet_accuracy": 0.85,
+      "eval_pet_f1": 0.7931034482758621,
+      "eval_pet_loss": 0.3718249797821045,
+      "eval_pet_macro_f1": 0.8377281947261663,
+      "eval_pet_micro_f1": 0.85,
+      "eval_pet_precision": 0.777027027027027,
+      "eval_pet_recall": 0.8098591549295775,
+      "eval_pet_runtime": 0.4379,
+      "eval_pet_samples_per_second": 913.543,
+      "eval_pet_specificity": 0.872093023255814,
+      "eval_pet_steps_per_second": 2.284,
+      "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:e32dc27c557e0f441bcd20b19146da9bd1fd47ccb03d8413c22d107b236c2a36
+size 5432