last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0179640718562875,
  "eval_steps": 9,
  "global_step": 42,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.023952095808383235,
      "eval_loss": 0.9337193965911865,
      "eval_runtime": 1.3258,
      "eval_samples_per_second": 13.576,
      "eval_steps_per_second": 6.788,
      "step": 1
    },
    {
      "epoch": 0.11976047904191617,
      "grad_norm": 3.728759288787842,
      "learning_rate": 5e-05,
      "loss": 0.8616,
      "step": 5
    },
    {
      "epoch": 0.2155688622754491,
      "eval_loss": 0.24477338790893555,
      "eval_runtime": 0.6155,
      "eval_samples_per_second": 29.246,
      "eval_steps_per_second": 14.623,
      "step": 9
    },
    {
      "epoch": 0.23952095808383234,
      "grad_norm": 3.8120970726013184,
      "learning_rate": 0.0001,
      "loss": 0.5099,
      "step": 10
    },
    {
      "epoch": 0.3592814371257485,
      "grad_norm": 0.3222445845603943,
      "learning_rate": 9.409606321741775e-05,
      "loss": 0.0491,
      "step": 15
    },
    {
      "epoch": 0.4311377245508982,
      "eval_loss": 0.0020451960153877735,
      "eval_runtime": 0.6164,
      "eval_samples_per_second": 29.203,
      "eval_steps_per_second": 14.601,
      "step": 18
    },
    {
      "epoch": 0.47904191616766467,
      "grad_norm": 0.08267904818058014,
      "learning_rate": 7.777851165098012e-05,
      "loss": 0.0037,
      "step": 20
    },
    {
      "epoch": 0.5988023952095808,
      "grad_norm": 0.023159213364124298,
      "learning_rate": 5.490085701647805e-05,
      "loss": 0.0013,
      "step": 25
    },
    {
      "epoch": 0.6467065868263473,
      "eval_loss": 0.0011916988296434283,
      "eval_runtime": 0.6209,
      "eval_samples_per_second": 28.992,
      "eval_steps_per_second": 14.496,
      "step": 27
    },
    {
      "epoch": 0.718562874251497,
      "grad_norm": 0.03186572715640068,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0013,
      "step": 30
    },
    {
      "epoch": 0.8383233532934131,
      "grad_norm": 0.023674650117754936,
      "learning_rate": 1.134947733186315e-05,
      "loss": 0.0011,
      "step": 35
    },
    {
      "epoch": 0.8622754491017964,
      "eval_loss": 0.0010952703887596726,
      "eval_runtime": 0.616,
      "eval_samples_per_second": 29.221,
      "eval_steps_per_second": 14.61,
      "step": 36
    },
    {
      "epoch": 0.9580838323353293,
      "grad_norm": 0.020750371739268303,
      "learning_rate": 9.607359798384785e-07,
      "loss": 0.0013,
      "step": 40
    }
  ],
  "logging_steps": 5,
  "max_steps": 42,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 11,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 373960287977472.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}