last-checkpoint/trainer_state.json

{
  "best_metric": 1.9664812088012695,
  "best_model_checkpoint": "./output/checkpoint-150",
  "epoch": 0.03235722375020223,
  "eval_steps": 150,
  "global_step": 150,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0021571482500134824,
      "grad_norm": 46.34773635864258,
      "learning_rate": 2.82842712474619e-06,
      "loss": 18.4687,
      "step": 10
    },
    {
      "epoch": 0.004314296500026965,
      "grad_norm": 43.269718170166016,
      "learning_rate": 5.65685424949238e-06,
      "loss": 18.0905,
      "step": 20
    },
    {
      "epoch": 0.006471444750040446,
      "grad_norm": 42.11697769165039,
      "learning_rate": 8.48528137423857e-06,
      "loss": 17.6192,
      "step": 30
    },
    {
      "epoch": 0.00862859300005393,
      "grad_norm": 36.54969024658203,
      "learning_rate": 1.131370849898476e-05,
      "loss": 17.4029,
      "step": 40
    },
    {
      "epoch": 0.010785741250067411,
      "grad_norm": 35.50069808959961,
      "learning_rate": 1.414213562373095e-05,
      "loss": 6.7145,
      "step": 50
    },
    {
      "epoch": 0.012942889500080892,
      "grad_norm": 59.332000732421875,
      "learning_rate": 1.697056274847714e-05,
      "loss": 6.775,
      "step": 60
    },
    {
      "epoch": 0.015100037750094376,
      "grad_norm": 60.65806579589844,
      "learning_rate": 1.979898987322333e-05,
      "loss": 18.2243,
      "step": 70
    },
    {
      "epoch": 0.01725718600010786,
      "grad_norm": 39.01629638671875,
      "learning_rate": 2.262741699796952e-05,
      "loss": 16.7181,
      "step": 80
    },
    {
      "epoch": 0.01941433425012134,
      "grad_norm": 46.44580078125,
      "learning_rate": 2.545584412271571e-05,
      "loss": 16.5121,
      "step": 90
    },
    {
      "epoch": 0.021571482500134822,
      "grad_norm": 40.75702667236328,
      "learning_rate": 2.82842712474619e-05,
      "loss": 16.5158,
      "step": 100
    },
    {
      "epoch": 0.023728630750148305,
      "grad_norm": 43.35206604003906,
      "learning_rate": 3.111269837220809e-05,
      "loss": 16.0931,
      "step": 110
    },
    {
      "epoch": 0.025885779000161785,
      "grad_norm": 43.914920806884766,
      "learning_rate": 3.394112549695428e-05,
      "loss": 16.1211,
      "step": 120
    },
    {
      "epoch": 0.028042927250175268,
      "grad_norm": 42.83036422729492,
      "learning_rate": 3.676955262170047e-05,
      "loss": 16.039,
      "step": 130
    },
    {
      "epoch": 0.03020007550018875,
      "grad_norm": 48.97016906738281,
      "learning_rate": 3.959797974644666e-05,
      "loss": 15.4753,
      "step": 140
    },
    {
      "epoch": 0.03235722375020223,
      "grad_norm": 40.86469268798828,
      "learning_rate": 4.242640687119285e-05,
      "loss": 15.3947,
      "step": 150
    },
    {
      "epoch": 0.03235722375020223,
      "eval_loss": 1.9664812088012695,
      "eval_runtime": 55.7926,
      "eval_samples_per_second": 8.962,
      "eval_steps_per_second": 8.962,
      "step": 150
    }
  ],
  "logging_steps": 10,
  "max_steps": 50000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 11,
  "save_steps": 150,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8441094036842086e+17,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}