trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.42855626344680786,
  "best_model_checkpoint": "bert_base_rand_20_v1_sst2/checkpoint-264",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1584,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 8.955648422241211,
      "learning_rate": 4.9e-05,
      "loss": 0.3865,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8119266055045872,
      "eval_loss": 0.42855626344680786,
      "eval_runtime": 0.5679,
      "eval_samples_per_second": 1535.473,
      "eval_steps_per_second": 7.043,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.470305442810059,
      "learning_rate": 4.8e-05,
      "loss": 0.2227,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.805045871559633,
      "eval_loss": 0.5209893584251404,
      "eval_runtime": 0.5602,
      "eval_samples_per_second": 1556.565,
      "eval_steps_per_second": 7.14,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.432986736297607,
      "learning_rate": 4.7e-05,
      "loss": 0.1634,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7855504587155964,
      "eval_loss": 0.5700969099998474,
      "eval_runtime": 0.5647,
      "eval_samples_per_second": 1544.11,
      "eval_steps_per_second": 7.083,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.539319038391113,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1285,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7935779816513762,
      "eval_loss": 0.5970008373260498,
      "eval_runtime": 0.5603,
      "eval_samples_per_second": 1556.348,
      "eval_steps_per_second": 7.139,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 9.496631622314453,
      "learning_rate": 4.5e-05,
      "loss": 0.1009,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8004587155963303,
      "eval_loss": 0.6181496381759644,
      "eval_runtime": 0.5599,
      "eval_samples_per_second": 1557.435,
      "eval_steps_per_second": 7.144,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 7.158674240112305,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0793,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.805045871559633,
      "eval_loss": 0.7211170196533203,
      "eval_runtime": 0.5656,
      "eval_samples_per_second": 1541.688,
      "eval_steps_per_second": 7.072,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 5.316079940232192e+16,
      "train_loss": 0.18020708512778233,
      "train_runtime": 663.5721,
      "train_samples_per_second": 5074.731,
      "train_steps_per_second": 19.892
    }
  ],
  "logging_steps": 1,
  "max_steps": 13200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 5
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.316079940232192e+16,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}