last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0031852205765249245,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00031852205765249244,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 1.3488,
      "step": 1
    },
    {
      "epoch": 0.00031852205765249244,
      "eval_loss": 2.888601303100586,
      "eval_runtime": 105.6611,
      "eval_samples_per_second": 6.256,
      "eval_steps_per_second": 3.133,
      "step": 1
    },
    {
      "epoch": 0.0006370441153049849,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 2.7731,
      "step": 2
    },
    {
      "epoch": 0.0009555661729574773,
      "grad_norm": Infinity,
      "learning_rate": 0.00015,
      "loss": 2.7692,
      "step": 3
    },
    {
      "epoch": 0.0009555661729574773,
      "eval_loss": 2.888601303100586,
      "eval_runtime": 105.6842,
      "eval_samples_per_second": 6.254,
      "eval_steps_per_second": 3.132,
      "step": 3
    },
    {
      "epoch": 0.0012740882306099697,
      "grad_norm": Infinity,
      "learning_rate": 0.0002,
      "loss": 1.3516,
      "step": 4
    },
    {
      "epoch": 0.0015926102882624622,
      "grad_norm": Infinity,
      "learning_rate": 0.00025,
      "loss": 3.2038,
      "step": 5
    },
    {
      "epoch": 0.0019111323459149545,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 2.3316,
      "step": 6
    },
    {
      "epoch": 0.0019111323459149545,
      "eval_loss": 2.888601303100586,
      "eval_runtime": 105.848,
      "eval_samples_per_second": 6.245,
      "eval_steps_per_second": 3.127,
      "step": 6
    },
    {
      "epoch": 0.002229654403567447,
      "grad_norm": Infinity,
      "learning_rate": 0.00035,
      "loss": 2.3754,
      "step": 7
    },
    {
      "epoch": 0.0025481764612199395,
      "grad_norm": Infinity,
      "learning_rate": 0.0004,
      "loss": 3.8853,
      "step": 8
    },
    {
      "epoch": 0.002866698518872432,
      "grad_norm": Infinity,
      "learning_rate": 0.00045000000000000004,
      "loss": 1.8824,
      "step": 9
    },
    {
      "epoch": 0.002866698518872432,
      "eval_loss": 2.888601303100586,
      "eval_runtime": 105.7148,
      "eval_samples_per_second": 6.253,
      "eval_steps_per_second": 3.131,
      "step": 9
    },
    {
      "epoch": 0.0031852205765249245,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 1.3865,
      "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": 1029368772034560.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}