Training in progress, epoch 4, checkpoint

checkpoint-400/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-400/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ab0c6896b1f143c60a154815903e69b60acff5d742dda6fce46160aa03ab980
+size 556799560

checkpoint-400/trainer_state.json

@@ -0,0 +1,217 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 50,
+  "global_step": 400,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5,
+      "grad_norm": 7.576278209686279,
+      "learning_rate": 1.8333333333333333e-05,
+      "loss": 0.5455,
+      "step": 50
+    },
+    {
+      "epoch": 0.5,
+      "eval_pet_AUC": 0.8744411211495174,
+      "eval_pet_accuracy": 0.7875,
+      "eval_pet_f1": 0.6413502109704642,
+      "eval_pet_loss": 0.40357428789138794,
+      "eval_pet_macro_f1": 0.7451866507783049,
+      "eval_pet_micro_f1": 0.7875,
+      "eval_pet_precision": 0.6129032258064516,
+      "eval_pet_recall": 0.672566371681416,
+      "eval_pet_runtime": 0.4329,
+      "eval_pet_samples_per_second": 924.048,
+      "eval_pet_specificity": 0.8327526132404182,
+      "eval_pet_steps_per_second": 2.31,
+      "step": 50
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 7.2780537605285645,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.3767,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "eval_pet_AUC": 0.9009589590206901,
+      "eval_pet_accuracy": 0.81,
+      "eval_pet_f1": 0.6935483870967742,
+      "eval_pet_loss": 0.3708844780921936,
+      "eval_pet_macro_f1": 0.7779336138382422,
+      "eval_pet_micro_f1": 0.81,
+      "eval_pet_precision": 0.6370370370370371,
+      "eval_pet_recall": 0.7610619469026548,
+      "eval_pet_runtime": 0.4373,
+      "eval_pet_samples_per_second": 914.609,
+      "eval_pet_specificity": 0.8292682926829268,
+      "eval_pet_steps_per_second": 2.287,
+      "step": 100
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 4.364447116851807,
+      "learning_rate": 1.5000000000000002e-05,
+      "loss": 0.2847,
+      "step": 150
+    },
+    {
+      "epoch": 1.5,
+      "eval_pet_AUC": 0.9033948999414141,
+      "eval_pet_accuracy": 0.8225,
+      "eval_pet_f1": 0.7171314741035857,
+      "eval_pet_loss": 0.4332599937915802,
+      "eval_pet_macro_f1": 0.7939027133723757,
+      "eval_pet_micro_f1": 0.8225,
+      "eval_pet_precision": 0.6521739130434783,
+      "eval_pet_recall": 0.7964601769911505,
+      "eval_pet_runtime": 0.436,
+      "eval_pet_samples_per_second": 917.428,
+      "eval_pet_specificity": 0.8327526132404182,
+      "eval_pet_steps_per_second": 2.294,
+      "step": 150
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 10.42080020904541,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.2694,
+      "step": 200
+    },
+    {
+      "epoch": 2.0,
+      "eval_pet_AUC": 0.8989238691375535,
+      "eval_pet_accuracy": 0.84,
+      "eval_pet_f1": 0.6952380952380952,
+      "eval_pet_loss": 0.4126374125480652,
+      "eval_pet_macro_f1": 0.7933817594834545,
+      "eval_pet_micro_f1": 0.84,
+      "eval_pet_precision": 0.7525773195876289,
+      "eval_pet_recall": 0.6460176991150443,
+      "eval_pet_runtime": 0.4605,
+      "eval_pet_samples_per_second": 868.692,
+      "eval_pet_specificity": 0.9163763066202091,
+      "eval_pet_steps_per_second": 2.172,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 13.263297080993652,
+      "learning_rate": 1.1666666666666668e-05,
+      "loss": 0.157,
+      "step": 250
+    },
+    {
+      "epoch": 2.5,
+      "eval_pet_AUC": 0.9052141469581574,
+      "eval_pet_accuracy": 0.8425,
+      "eval_pet_f1": 0.6956521739130435,
+      "eval_pet_loss": 0.5520845055580139,
+      "eval_pet_macro_f1": 0.7947063567710242,
+      "eval_pet_micro_f1": 0.8425,
+      "eval_pet_precision": 0.7659574468085106,
+      "eval_pet_recall": 0.6371681415929203,
+      "eval_pet_runtime": 0.4377,
+      "eval_pet_samples_per_second": 913.907,
+      "eval_pet_specificity": 0.9233449477351916,
+      "eval_pet_steps_per_second": 2.285,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 15.089917182922363,
+      "learning_rate": 1e-05,
+      "loss": 0.1704,
+      "step": 300
+    },
+    {
+      "epoch": 3.0,
+      "eval_pet_AUC": 0.9130769942339121,
+      "eval_pet_accuracy": 0.825,
+      "eval_pet_f1": 0.7265625,
+      "eval_pet_loss": 0.6370885968208313,
+      "eval_pet_macro_f1": 0.7989430147058824,
+      "eval_pet_micro_f1": 0.825,
+      "eval_pet_precision": 0.6503496503496503,
+      "eval_pet_recall": 0.8230088495575221,
+      "eval_pet_runtime": 0.4271,
+      "eval_pet_samples_per_second": 936.455,
+      "eval_pet_specificity": 0.8257839721254355,
+      "eval_pet_steps_per_second": 2.341,
+      "step": 300
+    },
+    {
+      "epoch": 3.5,
+      "grad_norm": 0.6412437558174133,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.1028,
+      "step": 350
+    },
+    {
+      "epoch": 3.5,
+      "eval_pet_AUC": 0.9160987943634178,
+      "eval_pet_accuracy": 0.8475,
+      "eval_pet_f1": 0.7336244541484717,
+      "eval_pet_loss": 0.663390040397644,
+      "eval_pet_macro_f1": 0.8133971657782639,
+      "eval_pet_micro_f1": 0.8475,
+      "eval_pet_precision": 0.7241379310344828,
+      "eval_pet_recall": 0.7433628318584071,
+      "eval_pet_runtime": 0.441,
+      "eval_pet_samples_per_second": 906.956,
+      "eval_pet_specificity": 0.8885017421602788,
+      "eval_pet_steps_per_second": 2.267,
+      "step": 350
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 4.150225639343262,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.1162,
+      "step": 400
+    },
+    {
+      "epoch": 4.0,
+      "eval_pet_AUC": 0.9196139496161081,
+      "eval_pet_accuracy": 0.8375,
+      "eval_pet_f1": 0.7430830039525692,
+      "eval_pet_loss": 0.7081860303878784,
+      "eval_pet_macro_f1": 0.8121265111170524,
+      "eval_pet_micro_f1": 0.8375,
+      "eval_pet_precision": 0.6714285714285714,
+      "eval_pet_recall": 0.831858407079646,
+      "eval_pet_runtime": 0.4313,
+      "eval_pet_samples_per_second": 927.468,
+      "eval_pet_specificity": 0.8397212543554007,
+      "eval_pet_steps_per_second": 2.319,
+      "step": 400
+    }
+  ],
+  "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": 153298799616000.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-400/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:75165ed46ad9ff161917859a5f4064580d3dc749becc3cf672a988710e00d3bc
+size 5432