last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0003470715835140998,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.470715835140998e-05,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 3.5797,
      "step": 1
    },
    {
      "epoch": 3.470715835140998e-05,
      "eval_loss": NaN,
      "eval_runtime": 255.3165,
      "eval_samples_per_second": 47.517,
      "eval_steps_per_second": 23.759,
      "step": 1
    },
    {
      "epoch": 6.941431670281996e-05,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 2
    },
    {
      "epoch": 0.00010412147505422994,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.00010412147505422994,
      "eval_loss": NaN,
      "eval_runtime": 255.1442,
      "eval_samples_per_second": 47.55,
      "eval_steps_per_second": 23.775,
      "step": 3
    },
    {
      "epoch": 0.00013882863340563992,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.0001735357917570499,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.00020824295010845987,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.00020824295010845987,
      "eval_loss": NaN,
      "eval_runtime": 256.6448,
      "eval_samples_per_second": 47.272,
      "eval_steps_per_second": 23.636,
      "step": 6
    },
    {
      "epoch": 0.00024295010845986985,
      "grad_norm": NaN,
      "learning_rate": 0.00014,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.00027765726681127983,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.0003123644251626898,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0003123644251626898,
      "eval_loss": NaN,
      "eval_runtime": 256.5453,
      "eval_samples_per_second": 47.29,
      "eval_steps_per_second": 23.645,
      "step": 9
    },
    {
      "epoch": 0.0003470715835140998,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "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
}