last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.007136485280999108,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007136485280999108,
      "grad_norm": 0.2850506603717804,
      "learning_rate": 2e-05,
      "loss": 1.4761,
      "step": 1
    },
    {
      "epoch": 0.0007136485280999108,
      "eval_loss": 0.3800647556781769,
      "eval_runtime": 62.7892,
      "eval_samples_per_second": 9.397,
      "eval_steps_per_second": 4.698,
      "step": 1
    },
    {
      "epoch": 0.0014272970561998216,
      "grad_norm": 0.2314898520708084,
      "learning_rate": 4e-05,
      "loss": 1.2496,
      "step": 2
    },
    {
      "epoch": 0.0021409455842997323,
      "grad_norm": 0.22756783664226532,
      "learning_rate": 6e-05,
      "loss": 1.2566,
      "step": 3
    },
    {
      "epoch": 0.0021409455842997323,
      "eval_loss": 0.3796400725841522,
      "eval_runtime": 62.9134,
      "eval_samples_per_second": 9.378,
      "eval_steps_per_second": 4.689,
      "step": 3
    },
    {
      "epoch": 0.002854594112399643,
      "grad_norm": 0.3028983771800995,
      "learning_rate": 8e-05,
      "loss": 1.2691,
      "step": 4
    },
    {
      "epoch": 0.003568242640499554,
      "grad_norm": 0.325924277305603,
      "learning_rate": 0.0001,
      "loss": 1.5082,
      "step": 5
    },
    {
      "epoch": 0.004281891168599465,
      "grad_norm": 0.33345407247543335,
      "learning_rate": 0.00012,
      "loss": 1.5615,
      "step": 6
    },
    {
      "epoch": 0.004281891168599465,
      "eval_loss": 0.3730396628379822,
      "eval_runtime": 62.943,
      "eval_samples_per_second": 9.374,
      "eval_steps_per_second": 4.687,
      "step": 6
    },
    {
      "epoch": 0.0049955396966993755,
      "grad_norm": 0.3155543804168701,
      "learning_rate": 0.00014,
      "loss": 1.2529,
      "step": 7
    },
    {
      "epoch": 0.005709188224799286,
      "grad_norm": 0.3585043251514435,
      "learning_rate": 0.00016,
      "loss": 1.1556,
      "step": 8
    },
    {
      "epoch": 0.006422836752899197,
      "grad_norm": 0.36284977197647095,
      "learning_rate": 0.00018,
      "loss": 1.3458,
      "step": 9
    },
    {
      "epoch": 0.006422836752899197,
      "eval_loss": 0.3418372571468353,
      "eval_runtime": 63.0822,
      "eval_samples_per_second": 9.353,
      "eval_steps_per_second": 4.676,
      "step": 9
    },
    {
      "epoch": 0.007136485280999108,
      "grad_norm": 0.36937472224235535,
      "learning_rate": 0.0002,
      "loss": 1.6261,
      "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": 6794734731264000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}