last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.011841326228537596,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0011841326228537595,
      "grad_norm": 1.09139883518219,
      "learning_rate": 2e-05,
      "loss": 2.1513,
      "step": 1
    },
    {
      "epoch": 0.0011841326228537595,
      "eval_loss": 1.6193398237228394,
      "eval_runtime": 7.3,
      "eval_samples_per_second": 48.767,
      "eval_steps_per_second": 24.384,
      "step": 1
    },
    {
      "epoch": 0.002368265245707519,
      "grad_norm": 0.7143899202346802,
      "learning_rate": 4e-05,
      "loss": 0.7548,
      "step": 2
    },
    {
      "epoch": 0.003552397868561279,
      "grad_norm": 0.8059898018836975,
      "learning_rate": 6e-05,
      "loss": 1.689,
      "step": 3
    },
    {
      "epoch": 0.003552397868561279,
      "eval_loss": 1.6181666851043701,
      "eval_runtime": 6.5058,
      "eval_samples_per_second": 54.72,
      "eval_steps_per_second": 27.36,
      "step": 3
    },
    {
      "epoch": 0.004736530491415038,
      "grad_norm": 1.1075034141540527,
      "learning_rate": 8e-05,
      "loss": 1.718,
      "step": 4
    },
    {
      "epoch": 0.005920663114268798,
      "grad_norm": 0.7382609844207764,
      "learning_rate": 0.0001,
      "loss": 1.5268,
      "step": 5
    },
    {
      "epoch": 0.007104795737122558,
      "grad_norm": 1.0442874431610107,
      "learning_rate": 0.00012,
      "loss": 1.1511,
      "step": 6
    },
    {
      "epoch": 0.007104795737122558,
      "eval_loss": 1.602021336555481,
      "eval_runtime": 6.9192,
      "eval_samples_per_second": 51.451,
      "eval_steps_per_second": 25.725,
      "step": 6
    },
    {
      "epoch": 0.008288928359976317,
      "grad_norm": 0.8361874222755432,
      "learning_rate": 0.00014,
      "loss": 1.8913,
      "step": 7
    },
    {
      "epoch": 0.009473060982830076,
      "grad_norm": 0.7909893989562988,
      "learning_rate": 0.00016,
      "loss": 1.813,
      "step": 8
    },
    {
      "epoch": 0.010657193605683837,
      "grad_norm": 0.801120400428772,
      "learning_rate": 0.00018,
      "loss": 1.7999,
      "step": 9
    },
    {
      "epoch": 0.010657193605683837,
      "eval_loss": 1.5673933029174805,
      "eval_runtime": 6.7071,
      "eval_samples_per_second": 53.078,
      "eval_steps_per_second": 26.539,
      "step": 9
    },
    {
      "epoch": 0.011841326228537596,
      "grad_norm": 1.0468965768814087,
      "learning_rate": 0.0002,
      "loss": 2.0179,
      "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": 133557245706240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}