trainer_state.json

{
  "best_metric": 0.38255795104555035,
  "best_model_checkpoint": "./step_test_microsoft_dit/checkpoint-2000",
  "epoch": 2.271580010095911,
  "eval_steps": 500,
  "global_step": 4500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2523977788995457,
      "eval_f1": 0.3040393637715986,
      "eval_loss": 0.16232739388942719,
      "eval_runtime": 573.4246,
      "eval_samples_per_second": 359.702,
      "eval_steps_per_second": 2.811,
      "step": 500
    },
    {
      "epoch": 0.5047955577990914,
      "grad_norm": 3.499274253845215,
      "learning_rate": 0.0004986304738420684,
      "loss": 0.1656,
      "step": 1000
    },
    {
      "epoch": 0.5047955577990914,
      "eval_f1": NaN,
      "eval_loss": 0.43072596192359924,
      "eval_runtime": 571.3845,
      "eval_samples_per_second": 360.986,
      "eval_steps_per_second": 2.821,
      "step": 1000
    },
    {
      "epoch": 0.7571933366986371,
      "eval_f1": NaN,
      "eval_loss": 0.14925973117351532,
      "eval_runtime": 574.5622,
      "eval_samples_per_second": 358.99,
      "eval_steps_per_second": 2.806,
      "step": 1500
    },
    {
      "epoch": 1.0095911155981827,
      "grad_norm": 0.47774481773376465,
      "learning_rate": 0.0004945369001834514,
      "loss": 0.14,
      "step": 2000
    },
    {
      "epoch": 1.0095911155981827,
      "eval_f1": 0.38255795104555035,
      "eval_loss": 0.18703292310237885,
      "eval_runtime": 575.9325,
      "eval_samples_per_second": 358.136,
      "eval_steps_per_second": 2.799,
      "step": 2000
    },
    {
      "epoch": 1.2619888944977284,
      "eval_f1": NaN,
      "eval_loss": 0.10761596262454987,
      "eval_runtime": 570.7076,
      "eval_samples_per_second": 361.414,
      "eval_steps_per_second": 2.825,
      "step": 2500
    },
    {
      "epoch": 1.514386673397274,
      "grad_norm": 0.8171533942222595,
      "learning_rate": 0.0004877641290737884,
      "loss": 0.1285,
      "step": 3000
    },
    {
      "epoch": 1.514386673397274,
      "eval_f1": 0.381203007518797,
      "eval_loss": 0.08960258215665817,
      "eval_runtime": 570.2026,
      "eval_samples_per_second": 361.735,
      "eval_steps_per_second": 2.827,
      "step": 3000
    },
    {
      "epoch": 1.76678445229682,
      "eval_f1": NaN,
      "eval_loss": 0.14047682285308838,
      "eval_runtime": 572.8668,
      "eval_samples_per_second": 360.052,
      "eval_steps_per_second": 2.814,
      "step": 3500
    },
    {
      "epoch": 2.0191822311963654,
      "grad_norm": 0.7625566124916077,
      "learning_rate": 0.0004783863644106502,
      "loss": 0.123,
      "step": 4000
    },
    {
      "epoch": 2.0191822311963654,
      "eval_f1": NaN,
      "eval_loss": 0.20614498853683472,
      "eval_runtime": 571.417,
      "eval_samples_per_second": 360.966,
      "eval_steps_per_second": 2.821,
      "step": 4000
    },
    {
      "epoch": 2.271580010095911,
      "eval_f1": 0.23949467373266953,
      "eval_loss": 0.22833816707134247,
      "eval_runtime": 576.5663,
      "eval_samples_per_second": 357.742,
      "eval_steps_per_second": 2.796,
      "step": 4500
    },
    {
      "epoch": 2.271580010095911,
      "step": 4500,
      "total_flos": 4.4623040438986555e+19,
      "train_loss": 0.13670311906602647,
      "train_runtime": 12745.6466,
      "train_samples_per_second": 301.279,
      "train_steps_per_second": 2.354
    }
  ],
  "logging_steps": 1000,
  "max_steps": 30000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 16,
  "save_steps": 500,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.4623040438986555e+19,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": null
}