trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.6535947712418301,
  "eval_steps": 25,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16339869281045752,
      "grad_norm": 2.9138197898864746,
      "learning_rate": 1.893939393939394e-05,
      "loss": 1.3885,
      "step": 25
    },
    {
      "epoch": 0.16339869281045752,
      "eval_loss": 0.6594953536987305,
      "eval_runtime": 633.7486,
      "eval_samples_per_second": 0.483,
      "eval_steps_per_second": 0.062,
      "step": 25
    },
    {
      "epoch": 0.32679738562091504,
      "grad_norm": 2.861112117767334,
      "learning_rate": 1.2626262626262628e-05,
      "loss": 0.6223,
      "step": 50
    },
    {
      "epoch": 0.32679738562091504,
      "eval_loss": 0.6024950742721558,
      "eval_runtime": 634.1943,
      "eval_samples_per_second": 0.483,
      "eval_steps_per_second": 0.061,
      "step": 50
    },
    {
      "epoch": 0.49019607843137253,
      "grad_norm": 4.142775058746338,
      "learning_rate": 6.313131313131314e-06,
      "loss": 0.6025,
      "step": 75
    },
    {
      "epoch": 0.49019607843137253,
      "eval_loss": 0.5404363870620728,
      "eval_runtime": 634.4774,
      "eval_samples_per_second": 0.482,
      "eval_steps_per_second": 0.061,
      "step": 75
    },
    {
      "epoch": 0.6535947712418301,
      "grad_norm": 14.336810111999512,
      "learning_rate": 0.0,
      "loss": 0.5024,
      "step": 100
    },
    {
      "epoch": 0.6535947712418301,
      "eval_loss": 0.5124824643135071,
      "eval_runtime": 635.0592,
      "eval_samples_per_second": 0.482,
      "eval_steps_per_second": 0.061,
      "step": 100
    }
  ],
  "logging_steps": 25,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5140361035776000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}