last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.000827934510380229,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.27934510380229e-05,
      "grad_norm": 1.5623795986175537,
      "learning_rate": 2e-05,
      "loss": 2.8657,
      "step": 1
    },
    {
      "epoch": 8.27934510380229e-05,
      "eval_loss": 0.6027017831802368,
      "eval_runtime": 383.6162,
      "eval_samples_per_second": 13.258,
      "eval_steps_per_second": 6.629,
      "step": 1
    },
    {
      "epoch": 0.0001655869020760458,
      "grad_norm": 0.9546533823013306,
      "learning_rate": 4e-05,
      "loss": 2.1009,
      "step": 2
    },
    {
      "epoch": 0.00024838035311406867,
      "grad_norm": 1.9295176267623901,
      "learning_rate": 6e-05,
      "loss": 2.5716,
      "step": 3
    },
    {
      "epoch": 0.00024838035311406867,
      "eval_loss": 0.599661111831665,
      "eval_runtime": 384.3256,
      "eval_samples_per_second": 13.234,
      "eval_steps_per_second": 6.617,
      "step": 3
    },
    {
      "epoch": 0.0003311738041520916,
      "grad_norm": 1.495574951171875,
      "learning_rate": 8e-05,
      "loss": 2.3509,
      "step": 4
    },
    {
      "epoch": 0.0004139672551901145,
      "grad_norm": 1.435874342918396,
      "learning_rate": 0.0001,
      "loss": 2.236,
      "step": 5
    },
    {
      "epoch": 0.0004967607062281373,
      "grad_norm": 2.2076618671417236,
      "learning_rate": 0.00012,
      "loss": 3.0957,
      "step": 6
    },
    {
      "epoch": 0.0004967607062281373,
      "eval_loss": 0.5404369831085205,
      "eval_runtime": 383.6732,
      "eval_samples_per_second": 13.256,
      "eval_steps_per_second": 6.628,
      "step": 6
    },
    {
      "epoch": 0.0005795541572661603,
      "grad_norm": 2.1777093410491943,
      "learning_rate": 0.00014,
      "loss": 1.9883,
      "step": 7
    },
    {
      "epoch": 0.0006623476083041832,
      "grad_norm": 2.2748916149139404,
      "learning_rate": 0.00016,
      "loss": 1.9523,
      "step": 8
    },
    {
      "epoch": 0.000745141059342206,
      "grad_norm": 2.3257896900177,
      "learning_rate": 0.00018,
      "loss": 1.6897,
      "step": 9
    },
    {
      "epoch": 0.000745141059342206,
      "eval_loss": 0.3113290071487427,
      "eval_runtime": 384.3903,
      "eval_samples_per_second": 13.231,
      "eval_steps_per_second": 6.616,
      "step": 9
    },
    {
      "epoch": 0.000827934510380229,
      "grad_norm": 1.796519160270691,
      "learning_rate": 0.0002,
      "loss": 1.0919,
      "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": 4670180850401280.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}