last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.05179669796050502,
  "eval_steps": 5,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.005179669796050502,
      "grad_norm": NaN,
      "learning_rate": 1e-05,
      "loss": 0.0,
      "step": 1
    },
    {
      "epoch": 0.005179669796050502,
      "eval_loss": NaN,
      "eval_runtime": 16.0963,
      "eval_samples_per_second": 20.253,
      "eval_steps_per_second": 10.127,
      "step": 1
    },
    {
      "epoch": 0.010359339592101004,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 292.1607,
      "step": 2
    },
    {
      "epoch": 0.015539009388151505,
      "grad_norm": NaN,
      "learning_rate": 3e-05,
      "loss": 0.0,
      "step": 3
    },
    {
      "epoch": 0.020718679184202008,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 0.0,
      "step": 4
    },
    {
      "epoch": 0.02589834898025251,
      "grad_norm": NaN,
      "learning_rate": 5e-05,
      "loss": 0.0,
      "step": 5
    },
    {
      "epoch": 0.02589834898025251,
      "eval_loss": NaN,
      "eval_runtime": 2.1965,
      "eval_samples_per_second": 148.421,
      "eval_steps_per_second": 74.21,
      "step": 5
    },
    {
      "epoch": 0.03107801877630301,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.0,
      "step": 6
    },
    {
      "epoch": 0.036257688572353515,
      "grad_norm": NaN,
      "learning_rate": 7e-05,
      "loss": 0.0,
      "step": 7
    },
    {
      "epoch": 0.041437358368404016,
      "grad_norm": NaN,
      "learning_rate": 8e-05,
      "loss": 0.0,
      "step": 8
    },
    {
      "epoch": 0.04661702816445452,
      "grad_norm": NaN,
      "learning_rate": 9e-05,
      "loss": 0.0,
      "step": 9
    },
    {
      "epoch": 0.05179669796050502,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 0.0,
      "step": 10
    },
    {
      "epoch": 0.05179669796050502,
      "eval_loss": NaN,
      "eval_runtime": 2.2704,
      "eval_samples_per_second": 143.587,
      "eval_steps_per_second": 71.794,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 20,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 58043324694528.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}