last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08492569002123142,
  "eval_steps": 2,
  "global_step": 5,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.016985138004246284,
      "grad_norm": 1.1376922130584717,
      "learning_rate": 1e-05,
      "loss": 0.6858,
      "step": 1
    },
    {
      "epoch": 0.016985138004246284,
      "eval_loss": 0.7396222949028015,
      "eval_runtime": 75.8691,
      "eval_samples_per_second": 1.318,
      "eval_steps_per_second": 0.659,
      "step": 1
    },
    {
      "epoch": 0.03397027600849257,
      "grad_norm": 0.9416837692260742,
      "learning_rate": 2e-05,
      "loss": 0.6133,
      "step": 2
    },
    {
      "epoch": 0.03397027600849257,
      "eval_loss": 0.7282164692878723,
      "eval_runtime": 8.4524,
      "eval_samples_per_second": 11.831,
      "eval_steps_per_second": 5.915,
      "step": 2
    },
    {
      "epoch": 0.050955414012738856,
      "grad_norm": 0.9376155734062195,
      "learning_rate": 3e-05,
      "loss": 0.5849,
      "step": 3
    },
    {
      "epoch": 0.06794055201698514,
      "grad_norm": 0.9625425934791565,
      "learning_rate": 4e-05,
      "loss": 0.634,
      "step": 4
    },
    {
      "epoch": 0.06794055201698514,
      "eval_loss": 0.6936032176017761,
      "eval_runtime": 7.5992,
      "eval_samples_per_second": 13.159,
      "eval_steps_per_second": 6.58,
      "step": 4
    },
    {
      "epoch": 0.08492569002123142,
      "grad_norm": 0.9886196255683899,
      "learning_rate": 5e-05,
      "loss": 0.5667,
      "step": 5
    }
  ],
  "logging_steps": 1,
  "max_steps": 5,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 2,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5637719677992960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}