last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004311273981461522,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0004311273981461522,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 11.9398,
      "step": 1
    },
    {
      "epoch": 0.0004311273981461522,
      "eval_loss": 12.775936126708984,
      "eval_runtime": 33.8995,
      "eval_samples_per_second": 14.425,
      "eval_steps_per_second": 14.425,
      "step": 1
    },
    {
      "epoch": 0.0008622547962923044,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 12.5284,
      "step": 2
    },
    {
      "epoch": 0.0012933821944384565,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 13.1201,
      "step": 3
    },
    {
      "epoch": 0.0012933821944384565,
      "eval_loss": 12.775936126708984,
      "eval_runtime": 33.7393,
      "eval_samples_per_second": 14.493,
      "eval_steps_per_second": 14.493,
      "step": 3
    },
    {
      "epoch": 0.0017245095925846087,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 12.4958,
      "step": 4
    },
    {
      "epoch": 0.002155636990730761,
      "grad_norm": 15.573494911193848,
      "learning_rate": 2e-05,
      "loss": 11.6975,
      "step": 5
    },
    {
      "epoch": 0.002586764388876913,
      "grad_norm": 13.357664108276367,
      "learning_rate": 4e-05,
      "loss": 12.9256,
      "step": 6
    },
    {
      "epoch": 0.002586764388876913,
      "eval_loss": 12.67633056640625,
      "eval_runtime": 33.8853,
      "eval_samples_per_second": 14.431,
      "eval_steps_per_second": 14.431,
      "step": 6
    },
    {
      "epoch": 0.003017891787023065,
      "grad_norm": 11.505864143371582,
      "learning_rate": 6e-05,
      "loss": 13.1193,
      "step": 7
    },
    {
      "epoch": 0.0034490191851692174,
      "grad_norm": 15.953032493591309,
      "learning_rate": 8e-05,
      "loss": 11.2816,
      "step": 8
    },
    {
      "epoch": 0.0038801465833153696,
      "grad_norm": 18.366994857788086,
      "learning_rate": 0.0001,
      "loss": 11.6187,
      "step": 9
    },
    {
      "epoch": 0.0038801465833153696,
      "eval_loss": 9.702347755432129,
      "eval_runtime": 37.8232,
      "eval_samples_per_second": 12.929,
      "eval_steps_per_second": 12.929,
      "step": 9
    },
    {
      "epoch": 0.004311273981461522,
      "grad_norm": 21.4761905670166,
      "learning_rate": 0.00012,
      "loss": 10.1439,
      "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": 1849564248145920.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}