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 sha256:5b662d7eb9268bb2479a0169b87367f3dca15c75eb5f47b44199d7b291fdb61a
+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": 6.69125509262085,
+      "learning_rate": 2.75e-05,
+      "loss": 0.5067,
+      "step": 50
+    },
+    {
+      "epoch": 0.5,
+      "eval_pet_AUC": 0.8991702150889835,
+      "eval_pet_accuracy": 0.7875,
+      "eval_pet_f1": 0.6351931330472103,
+      "eval_pet_loss": 0.43754860758781433,
+      "eval_pet_macro_f1": 0.7426406582343634,
+      "eval_pet_micro_f1": 0.7875,
+      "eval_pet_precision": 0.8131868131868132,
+      "eval_pet_recall": 0.5211267605633803,
+      "eval_pet_runtime": 0.4448,
+      "eval_pet_samples_per_second": 899.31,
+      "eval_pet_specificity": 0.9341085271317829,
+      "eval_pet_steps_per_second": 2.248,
+      "step": 50
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 6.729673385620117,
+      "learning_rate": 2.5e-05,
+      "loss": 0.3894,
+      "step": 100
+    },
+    {
+      "epoch": 1.0,
+      "eval_pet_AUC": 0.9216344579102522,
+      "eval_pet_accuracy": 0.8375,
+      "eval_pet_f1": 0.7854785478547854,
+      "eval_pet_loss": 0.386077344417572,
+      "eval_pet_macro_f1": 0.8273469198026442,
+      "eval_pet_micro_f1": 0.8375,
+      "eval_pet_precision": 0.7391304347826086,
+      "eval_pet_recall": 0.8380281690140845,
+      "eval_pet_runtime": 0.4465,
+      "eval_pet_samples_per_second": 895.949,
+      "eval_pet_specificity": 0.8372093023255814,
+      "eval_pet_steps_per_second": 2.24,
+      "step": 100
+    },
+    {
+      "epoch": 1.5,
+      "grad_norm": 8.36129093170166,
+      "learning_rate": 2.25e-05,
+      "loss": 0.302,
+      "step": 150
+    },
+    {
+      "epoch": 1.5,
+      "eval_pet_AUC": 0.923190304618408,
+      "eval_pet_accuracy": 0.84,
+      "eval_pet_f1": 0.7922077922077922,
+      "eval_pet_loss": 0.3753361403942108,
+      "eval_pet_macro_f1": 0.8310632456973921,
+      "eval_pet_micro_f1": 0.84,
+      "eval_pet_precision": 0.7349397590361446,
+      "eval_pet_recall": 0.8591549295774648,
+      "eval_pet_runtime": 0.4462,
+      "eval_pet_samples_per_second": 896.529,
+      "eval_pet_specificity": 0.8294573643410853,
+      "eval_pet_steps_per_second": 2.241,
+      "step": 150
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 10.58817195892334,
+      "learning_rate": 1.9999999999999998e-05,
+      "loss": 0.2493,
+      "step": 200
+    },
+    {
+      "epoch": 2.0,
+      "eval_pet_AUC": 0.932702805983186,
+      "eval_pet_accuracy": 0.85,
+      "eval_pet_f1": 0.8064516129032258,
+      "eval_pet_loss": 0.4389405846595764,
+      "eval_pet_macro_f1": 0.8420013166556946,
+      "eval_pet_micro_f1": 0.85,
+      "eval_pet_precision": 0.7440476190476191,
+      "eval_pet_recall": 0.8802816901408451,
+      "eval_pet_runtime": 0.4511,
+      "eval_pet_samples_per_second": 886.77,
+      "eval_pet_specificity": 0.8333333333333334,
+      "eval_pet_steps_per_second": 2.217,
+      "step": 200
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 5.658120632171631,
+      "learning_rate": 1.7500000000000002e-05,
+      "loss": 0.168,
+      "step": 250
+    },
+    {
+      "epoch": 2.5,
+      "eval_pet_AUC": 0.9210612512283001,
+      "eval_pet_accuracy": 0.8325,
+      "eval_pet_f1": 0.7330677290836654,
+      "eval_pet_loss": 0.6072443723678589,
+      "eval_pet_macro_f1": 0.8055138281119603,
+      "eval_pet_micro_f1": 0.8325,
+      "eval_pet_precision": 0.8440366972477065,
+      "eval_pet_recall": 0.647887323943662,
+      "eval_pet_runtime": 0.4417,
+      "eval_pet_samples_per_second": 905.551,
+      "eval_pet_specificity": 0.9341085271317829,
+      "eval_pet_steps_per_second": 2.264,
+      "step": 250
+    },
+    {
+      "epoch": 3.0,
+      "grad_norm": 1.3053780794143677,
+      "learning_rate": 1.5e-05,
+      "loss": 0.1642,
+      "step": 300
+    },
+    {
+      "epoch": 3.0,
+      "eval_pet_AUC": 0.9298777159078503,
+      "eval_pet_accuracy": 0.8825,
+      "eval_pet_f1": 0.8362369337979094,
+      "eval_pet_loss": 0.4500144124031067,
+      "eval_pet_macro_f1": 0.8723095000373562,
+      "eval_pet_micro_f1": 0.8825,
+      "eval_pet_precision": 0.8275862068965517,
+      "eval_pet_recall": 0.8450704225352113,
+      "eval_pet_runtime": 0.4358,
+      "eval_pet_samples_per_second": 917.924,
+      "eval_pet_specificity": 0.9031007751937985,
+      "eval_pet_steps_per_second": 2.295,
+      "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 sha256:9ce127e8b9a9e625b848ff93238c56d0bbf79e8850eab8d7acd300b3ddfe61cc
+size 5432