last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0014123096912691015,
  "eval_steps": 4,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.649238765076406e-05,
      "grad_norm": NaN,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 5.649238765076406e-05,
      "eval_loss": NaN,
      "eval_runtime": 793.0271,
      "eval_samples_per_second": 9.399,
      "eval_steps_per_second": 2.35,
      "step": 1
    },
    {
      "epoch": 0.00011298477530152812,
      "grad_norm": NaN,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.00016947716295229217,
      "grad_norm": NaN,
      "learning_rate": 3e-06,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.00022596955060305625,
      "grad_norm": NaN,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.00022596955060305625,
      "eval_loss": NaN,
      "eval_runtime": 792.1663,
      "eval_samples_per_second": 9.41,
      "eval_steps_per_second": 2.353,
      "step": 4
    },
    {
      "epoch": 0.0002824619382538203,
      "grad_norm": NaN,
      "learning_rate": 5e-06,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00033895432590458433,
      "grad_norm": NaN,
      "learning_rate": 6e-06,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0003954467135553484,
      "grad_norm": NaN,
      "learning_rate": 7e-06,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.0004519391012061125,
      "grad_norm": NaN,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0004519391012061125,
      "eval_loss": NaN,
      "eval_runtime": 792.3101,
      "eval_samples_per_second": 9.408,
      "eval_steps_per_second": 2.353,
      "step": 8
    },
    {
      "epoch": 0.0005084314888568766,
      "grad_norm": NaN,
      "learning_rate": 9e-06,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0005649238765076406,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0006214162641584046,
      "grad_norm": NaN,
      "learning_rate": 9.890738003669029e-06,
      "loss": 0.0,
      "step": 11
    },
    {
      "epoch": 0.0006779086518091687,
      "grad_norm": NaN,
      "learning_rate": 9.567727288213005e-06,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0006779086518091687,
      "eval_loss": NaN,
      "eval_runtime": 792.3034,
      "eval_samples_per_second": 9.408,
      "eval_steps_per_second": 2.353,
      "step": 12
    },
    {
      "epoch": 0.0007344010394599328,
      "grad_norm": NaN,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.0,
      "step": 13
    },
    {
      "epoch": 0.0007908934271106968,
      "grad_norm": NaN,
      "learning_rate": 8.345653031794292e-06,
      "loss": 0.0,
      "step": 14
    },
    {
      "epoch": 0.0008473858147614609,
      "grad_norm": NaN,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.000903878202412225,
      "grad_norm": NaN,
      "learning_rate": 6.545084971874738e-06,
      "loss": 0.0,
      "step": 16
    },
    {
      "epoch": 0.000903878202412225,
      "eval_loss": NaN,
      "eval_runtime": 792.1187,
      "eval_samples_per_second": 9.41,
      "eval_steps_per_second": 2.353,
      "step": 16
    },
    {
      "epoch": 0.000960370590062989,
      "grad_norm": NaN,
      "learning_rate": 5.522642316338268e-06,
      "loss": 0.0,
      "step": 17
    },
    {
      "epoch": 0.0010168629777137532,
      "grad_norm": NaN,
      "learning_rate": 4.477357683661734e-06,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.001073355365364517,
      "grad_norm": NaN,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 0.0,
      "step": 19
    },
    {
      "epoch": 0.0011298477530152812,
      "grad_norm": NaN,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0011298477530152812,
      "eval_loss": NaN,
      "eval_runtime": 791.3598,
      "eval_samples_per_second": 9.419,
      "eval_steps_per_second": 2.355,
      "step": 20
    },
    {
      "epoch": 0.0011863401406660454,
      "grad_norm": NaN,
      "learning_rate": 1.6543469682057105e-06,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0012428325283168093,
      "grad_norm": NaN,
      "learning_rate": 9.549150281252633e-07,
      "loss": 0.0,
      "step": 22
    },
    {
      "epoch": 0.0012993249159675734,
      "grad_norm": NaN,
      "learning_rate": 4.322727117869951e-07,
      "loss": 0.0,
      "step": 23
    },
    {
      "epoch": 0.0013558173036183373,
      "grad_norm": NaN,
      "learning_rate": 1.0926199633097156e-07,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.0013558173036183373,
      "eval_loss": NaN,
      "eval_runtime": 791.728,
      "eval_samples_per_second": 9.415,
      "eval_steps_per_second": 2.354,
      "step": 24
    },
    {
      "epoch": 0.0014123096912691015,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 25,
  "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": 8240382974361600.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}