last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00013361793954456326,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.3361793954456326e-05,
      "grad_norm": 0.8661659955978394,
      "learning_rate": 2e-05,
      "loss": 1.3794,
      "step": 1
    },
    {
      "epoch": 1.3361793954456326e-05,
      "eval_loss": 1.3569523096084595,
      "eval_runtime": 3578.2788,
      "eval_samples_per_second": 8.806,
      "eval_steps_per_second": 4.403,
      "step": 1
    },
    {
      "epoch": 2.6723587908912652e-05,
      "grad_norm": 1.0288063287734985,
      "learning_rate": 4e-05,
      "loss": 1.3412,
      "step": 2
    },
    {
      "epoch": 4.0085381863368976e-05,
      "grad_norm": 0.7789365649223328,
      "learning_rate": 6e-05,
      "loss": 1.1374,
      "step": 3
    },
    {
      "epoch": 4.0085381863368976e-05,
      "eval_loss": 1.351515531539917,
      "eval_runtime": 958.916,
      "eval_samples_per_second": 32.862,
      "eval_steps_per_second": 16.431,
      "step": 3
    },
    {
      "epoch": 5.3447175817825303e-05,
      "grad_norm": 0.8093341588973999,
      "learning_rate": 8e-05,
      "loss": 1.1683,
      "step": 4
    },
    {
      "epoch": 6.680896977228163e-05,
      "grad_norm": 0.9298834800720215,
      "learning_rate": 0.0001,
      "loss": 1.3035,
      "step": 5
    },
    {
      "epoch": 8.017076372673795e-05,
      "grad_norm": 1.0221952199935913,
      "learning_rate": 0.00012,
      "loss": 1.1864,
      "step": 6
    },
    {
      "epoch": 8.017076372673795e-05,
      "eval_loss": 1.2463455200195312,
      "eval_runtime": 959.2892,
      "eval_samples_per_second": 32.849,
      "eval_steps_per_second": 16.425,
      "step": 6
    },
    {
      "epoch": 9.353255768119427e-05,
      "grad_norm": 1.2705566883087158,
      "learning_rate": 0.00014,
      "loss": 1.3585,
      "step": 7
    },
    {
      "epoch": 0.00010689435163565061,
      "grad_norm": 1.5416756868362427,
      "learning_rate": 0.00016,
      "loss": 1.0534,
      "step": 8
    },
    {
      "epoch": 0.00012025614559010693,
      "grad_norm": 1.3119838237762451,
      "learning_rate": 0.00018,
      "loss": 0.8679,
      "step": 9
    },
    {
      "epoch": 0.00012025614559010693,
      "eval_loss": 0.7938746213912964,
      "eval_runtime": 959.1171,
      "eval_samples_per_second": 32.855,
      "eval_steps_per_second": 16.428,
      "step": 9
    },
    {
      "epoch": 0.00013361793954456326,
      "grad_norm": 1.5242027044296265,
      "learning_rate": 0.0002,
      "loss": 0.6953,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}