last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.032480714575720666,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0032480714575720666,
      "grad_norm": 9.518817901611328,
      "learning_rate": 1e-05,
      "loss": 12.119,
      "step": 1
    },
    {
      "epoch": 0.0032480714575720666,
      "eval_loss": 12.267593383789062,
      "eval_runtime": 108.5306,
      "eval_samples_per_second": 4.782,
      "eval_steps_per_second": 2.396,
      "step": 1
    },
    {
      "epoch": 0.006496142915144133,
      "grad_norm": 10.461214065551758,
      "learning_rate": 2e-05,
      "loss": 12.6094,
      "step": 2
    },
    {
      "epoch": 0.0097442143727162,
      "grad_norm": 9.743880271911621,
      "learning_rate": 3e-05,
      "loss": 11.767,
      "step": 3
    },
    {
      "epoch": 0.0097442143727162,
      "eval_loss": 11.903098106384277,
      "eval_runtime": 24.449,
      "eval_samples_per_second": 21.228,
      "eval_steps_per_second": 10.634,
      "step": 3
    },
    {
      "epoch": 0.012992285830288266,
      "grad_norm": 10.506345748901367,
      "learning_rate": 4e-05,
      "loss": 11.9058,
      "step": 4
    },
    {
      "epoch": 0.016240357287860333,
      "grad_norm": 10.899349212646484,
      "learning_rate": 5e-05,
      "loss": 11.8389,
      "step": 5
    },
    {
      "epoch": 0.0194884287454324,
      "grad_norm": 11.90410041809082,
      "learning_rate": 6e-05,
      "loss": 11.3438,
      "step": 6
    },
    {
      "epoch": 0.0194884287454324,
      "eval_loss": 10.635560035705566,
      "eval_runtime": 25.6821,
      "eval_samples_per_second": 20.209,
      "eval_steps_per_second": 10.124,
      "step": 6
    },
    {
      "epoch": 0.022736500203004467,
      "grad_norm": 12.569297790527344,
      "learning_rate": 7e-05,
      "loss": 10.9902,
      "step": 7
    },
    {
      "epoch": 0.025984571660576532,
      "grad_norm": 15.150697708129883,
      "learning_rate": 8e-05,
      "loss": 9.8283,
      "step": 8
    },
    {
      "epoch": 0.029232643118148598,
      "grad_norm": 16.48304557800293,
      "learning_rate": 9e-05,
      "loss": 8.6592,
      "step": 9
    },
    {
      "epoch": 0.029232643118148598,
      "eval_loss": 6.93628454208374,
      "eval_runtime": 25.6766,
      "eval_samples_per_second": 20.213,
      "eval_steps_per_second": 10.126,
      "step": 9
    },
    {
      "epoch": 0.032480714575720666,
      "grad_norm": 18.78893280029297,
      "learning_rate": 0.0001,
      "loss": 6.8171,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7398256992583680.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}