last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.014545454545454545,
  "eval_steps": 8,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014545454545454545,
      "eval_loss": 0.405824214220047,
      "eval_runtime": 37.2045,
      "eval_samples_per_second": 7.795,
      "eval_steps_per_second": 3.897,
      "step": 1
    },
    {
      "epoch": 0.004363636363636364,
      "grad_norm": 0.4422492980957031,
      "learning_rate": 6e-05,
      "loss": 0.3811,
      "step": 3
    },
    {
      "epoch": 0.008727272727272728,
      "grad_norm": 0.5055964589118958,
      "learning_rate": 0.00012,
      "loss": 0.237,
      "step": 6
    },
    {
      "epoch": 0.011636363636363636,
      "eval_loss": 0.154419407248497,
      "eval_runtime": 37.2223,
      "eval_samples_per_second": 7.791,
      "eval_steps_per_second": 3.896,
      "step": 8
    },
    {
      "epoch": 0.01309090909090909,
      "grad_norm": 0.9195519089698792,
      "learning_rate": 0.00018,
      "loss": 0.2196,
      "step": 9
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3709436417802240.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}