last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004156563907170073,
  "eval_steps": 5,
  "global_step": 30,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00013855213023900243,
      "eval_loss": NaN,
      "eval_runtime": 375.3904,
      "eval_samples_per_second": 8.096,
      "eval_steps_per_second": 4.049,
      "step": 1
    },
    {
      "epoch": 0.0004156563907170073,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.0006927606511950121,
      "eval_loss": NaN,
      "eval_runtime": 376.7062,
      "eval_samples_per_second": 8.067,
      "eval_steps_per_second": 4.035,
      "step": 5
    },
    {
      "epoch": 0.0008313127814340146,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.0012469691721510219,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.0013855213023900243,
      "eval_loss": NaN,
      "eval_runtime": 376.4933,
      "eval_samples_per_second": 8.072,
      "eval_steps_per_second": 4.037,
      "step": 10
    },
    {
      "epoch": 0.0016626255628680291,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 0.0,
      "step": 12
    },
    {
      "epoch": 0.0020782819535850364,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 0.0,
      "step": 15
    },
    {
      "epoch": 0.0020782819535850364,
      "eval_loss": NaN,
      "eval_runtime": 376.4775,
      "eval_samples_per_second": 8.072,
      "eval_steps_per_second": 4.037,
      "step": 15
    },
    {
      "epoch": 0.0024939383443020437,
      "grad_norm": NaN,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0,
      "step": 18
    },
    {
      "epoch": 0.0027710426047800486,
      "eval_loss": NaN,
      "eval_runtime": 376.5179,
      "eval_samples_per_second": 8.071,
      "eval_steps_per_second": 4.037,
      "step": 20
    },
    {
      "epoch": 0.002909594735019051,
      "grad_norm": NaN,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0,
      "step": 21
    },
    {
      "epoch": 0.0033252511257360583,
      "grad_norm": NaN,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0,
      "step": 24
    },
    {
      "epoch": 0.0034638032559750607,
      "eval_loss": NaN,
      "eval_runtime": 376.5643,
      "eval_samples_per_second": 8.07,
      "eval_steps_per_second": 4.036,
      "step": 25
    },
    {
      "epoch": 0.0037409075164530656,
      "grad_norm": NaN,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0,
      "step": 27
    },
    {
      "epoch": 0.004156563907170073,
      "grad_norm": NaN,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 30
    },
    {
      "epoch": 0.004156563907170073,
      "eval_loss": NaN,
      "eval_runtime": 376.6164,
      "eval_samples_per_second": 8.069,
      "eval_steps_per_second": 4.036,
      "step": 30
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 15,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.85386137550848e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}