last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0005026072752403091,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 5.026072752403091e-05,
      "grad_norm": 2.132338762283325,
      "learning_rate": 2e-05,
      "loss": 2.5557,
      "step": 1
    },
    {
      "epoch": 5.026072752403091e-05,
      "eval_loss": 2.631843328475952,
      "eval_runtime": 1241.7974,
      "eval_samples_per_second": 6.747,
      "eval_steps_per_second": 3.373,
      "step": 1
    },
    {
      "epoch": 0.00010052145504806182,
      "grad_norm": 2.1546549797058105,
      "learning_rate": 4e-05,
      "loss": 2.7389,
      "step": 2
    },
    {
      "epoch": 0.00015078218257209272,
      "grad_norm": 2.0972418785095215,
      "learning_rate": 6e-05,
      "loss": 2.3617,
      "step": 3
    },
    {
      "epoch": 0.00015078218257209272,
      "eval_loss": 2.6196017265319824,
      "eval_runtime": 1305.1031,
      "eval_samples_per_second": 6.419,
      "eval_steps_per_second": 3.21,
      "step": 3
    },
    {
      "epoch": 0.00020104291009612364,
      "grad_norm": 1.7208608388900757,
      "learning_rate": 8e-05,
      "loss": 2.1952,
      "step": 4
    },
    {
      "epoch": 0.00025130363762015457,
      "grad_norm": 1.6549865007400513,
      "learning_rate": 0.0001,
      "loss": 2.5855,
      "step": 5
    },
    {
      "epoch": 0.00030156436514418544,
      "grad_norm": 2.5987608432769775,
      "learning_rate": 0.00012,
      "loss": 2.8653,
      "step": 6
    },
    {
      "epoch": 0.00030156436514418544,
      "eval_loss": 2.5160012245178223,
      "eval_runtime": 2098.8915,
      "eval_samples_per_second": 3.992,
      "eval_steps_per_second": 1.996,
      "step": 6
    },
    {
      "epoch": 0.00035182509266821636,
      "grad_norm": 1.7206662893295288,
      "learning_rate": 0.00014,
      "loss": 2.234,
      "step": 7
    },
    {
      "epoch": 0.0004020858201922473,
      "grad_norm": 1.7196815013885498,
      "learning_rate": 0.00016,
      "loss": 2.1952,
      "step": 8
    },
    {
      "epoch": 0.0004523465477162782,
      "grad_norm": 2.0840251445770264,
      "learning_rate": 0.00018,
      "loss": 2.5875,
      "step": 9
    },
    {
      "epoch": 0.0004523465477162782,
      "eval_loss": 2.3682916164398193,
      "eval_runtime": 1865.3751,
      "eval_samples_per_second": 4.491,
      "eval_steps_per_second": 2.246,
      "step": 9
    },
    {
      "epoch": 0.0005026072752403091,
      "grad_norm": 2.020665407180786,
      "learning_rate": 0.0002,
      "loss": 2.2747,
      "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": 360477150412800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}