last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0021749768908705344,
  "eval_steps": 10,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.349953781741069e-05,
      "eval_loss": NaN,
      "eval_runtime": 401.9814,
      "eval_samples_per_second": 24.081,
      "eval_steps_per_second": 12.04,
      "step": 1
    },
    {
      "epoch": 0.00021749768908705346,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.0004349953781741069,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.0004349953781741069,
      "eval_loss": NaN,
      "eval_runtime": 403.3545,
      "eval_samples_per_second": 23.999,
      "eval_steps_per_second": 11.999,
      "step": 10
    },
    {
      "epoch": 0.0006524930672611604,
      "grad_norm": NaN,
      "learning_rate": 9.619397662556435e-05,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0008699907563482138,
      "grad_norm": NaN,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.0,
      "step": 20
    },
    {
      "epoch": 0.0008699907563482138,
      "eval_loss": NaN,
      "eval_runtime": 405.5704,
      "eval_samples_per_second": 23.868,
      "eval_steps_per_second": 11.934,
      "step": 20
    },
    {
      "epoch": 0.0010874884454352672,
      "grad_norm": NaN,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.0,
      "step": 25
    },
    {
      "epoch": 0.0013049861345223207,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.0013049861345223207,
      "eval_loss": NaN,
      "eval_runtime": 401.1651,
      "eval_samples_per_second": 24.13,
      "eval_steps_per_second": 12.065,
      "step": 30
    },
    {
      "epoch": 0.001522483823609374,
      "grad_norm": NaN,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.0,
      "step": 35
    },
    {
      "epoch": 0.0017399815126964277,
      "grad_norm": NaN,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.0,
      "step": 40
    },
    {
      "epoch": 0.0017399815126964277,
      "eval_loss": NaN,
      "eval_runtime": 403.1526,
      "eval_samples_per_second": 24.011,
      "eval_steps_per_second": 12.005,
      "step": 40
    },
    {
      "epoch": 0.0019574792017834812,
      "grad_norm": NaN,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.0,
      "step": 45
    },
    {
      "epoch": 0.0021749768908705344,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 50
    },
    {
      "epoch": 0.0021749768908705344,
      "eval_loss": NaN,
      "eval_runtime": 409.0549,
      "eval_samples_per_second": 23.664,
      "eval_steps_per_second": 11.832,
      "step": 50
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2497096590557184.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}