trainer_state.json

{
  "best_global_step": 264,
  "best_metric": 0.42003896832466125,
  "best_model_checkpoint": "bert_base_rand_5_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": 8.597793579101562,
      "learning_rate": 4.9e-05,
      "loss": 0.3882,
      "step": 264
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8119266055045872,
      "eval_loss": 0.42003896832466125,
      "eval_runtime": 0.5611,
      "eval_samples_per_second": 1554.044,
      "eval_steps_per_second": 7.129,
      "step": 264
    },
    {
      "epoch": 2.0,
      "grad_norm": 8.21224308013916,
      "learning_rate": 4.8e-05,
      "loss": 0.2189,
      "step": 528
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8188073394495413,
      "eval_loss": 0.49200284481048584,
      "eval_runtime": 0.5653,
      "eval_samples_per_second": 1542.542,
      "eval_steps_per_second": 7.076,
      "step": 528
    },
    {
      "epoch": 3.0,
      "grad_norm": 8.396758079528809,
      "learning_rate": 4.7e-05,
      "loss": 0.1616,
      "step": 792
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7958715596330275,
      "eval_loss": 0.5462589263916016,
      "eval_runtime": 0.564,
      "eval_samples_per_second": 1546.142,
      "eval_steps_per_second": 7.092,
      "step": 792
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.590652942657471,
      "learning_rate": 4.600000000000001e-05,
      "loss": 0.1243,
      "step": 1056
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8153669724770642,
      "eval_loss": 0.5910820364952087,
      "eval_runtime": 0.5714,
      "eval_samples_per_second": 1526.112,
      "eval_steps_per_second": 7.001,
      "step": 1056
    },
    {
      "epoch": 5.0,
      "grad_norm": 9.480948448181152,
      "learning_rate": 4.5e-05,
      "loss": 0.0986,
      "step": 1320
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7958715596330275,
      "eval_loss": 0.6072150468826294,
      "eval_runtime": 0.57,
      "eval_samples_per_second": 1529.902,
      "eval_steps_per_second": 7.018,
      "step": 1320
    },
    {
      "epoch": 6.0,
      "grad_norm": 4.780208110809326,
      "learning_rate": 4.4000000000000006e-05,
      "loss": 0.0792,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8027522935779816,
      "eval_loss": 0.7792913317680359,
      "eval_runtime": 0.5642,
      "eval_samples_per_second": 1545.495,
      "eval_steps_per_second": 7.089,
      "step": 1584
    },
    {
      "epoch": 6.0,
      "step": 1584,
      "total_flos": 5.316079940232192e+16,
      "train_loss": 0.17848558859391647,
      "train_runtime": 661.7504,
      "train_samples_per_second": 5088.701,
      "train_steps_per_second": 19.947
    }
  ],
  "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
}