last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.000745141059342206,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.27934510380229e-05,
      "grad_norm": 1.4978296756744385,
      "learning_rate": 2e-05,
      "loss": 2.8657,
      "step": 1
    },
    {
      "epoch": 8.27934510380229e-05,
      "eval_loss": 0.6027017831802368,
      "eval_runtime": 386.7587,
      "eval_samples_per_second": 13.15,
      "eval_steps_per_second": 6.575,
      "step": 1
    },
    {
      "epoch": 0.0001655869020760458,
      "grad_norm": 0.8970440626144409,
      "learning_rate": 4e-05,
      "loss": 2.1009,
      "step": 2
    },
    {
      "epoch": 0.00024838035311406867,
      "grad_norm": 1.8807094097137451,
      "learning_rate": 6e-05,
      "loss": 2.5709,
      "step": 3
    },
    {
      "epoch": 0.00024838035311406867,
      "eval_loss": 0.5998329520225525,
      "eval_runtime": 363.0311,
      "eval_samples_per_second": 14.01,
      "eval_steps_per_second": 7.005,
      "step": 3
    },
    {
      "epoch": 0.0003311738041520916,
      "grad_norm": 1.4324417114257812,
      "learning_rate": 8e-05,
      "loss": 2.3556,
      "step": 4
    },
    {
      "epoch": 0.0004139672551901145,
      "grad_norm": 1.3803950548171997,
      "learning_rate": 0.0001,
      "loss": 2.2364,
      "step": 5
    },
    {
      "epoch": 0.0004967607062281373,
      "grad_norm": 4.585956573486328,
      "learning_rate": 0.00012,
      "loss": 3.0927,
      "step": 6
    },
    {
      "epoch": 0.0004967607062281373,
      "eval_loss": 0.5434311032295227,
      "eval_runtime": 387.3334,
      "eval_samples_per_second": 13.131,
      "eval_steps_per_second": 6.565,
      "step": 6
    },
    {
      "epoch": 0.0005795541572661603,
      "grad_norm": 2.122243881225586,
      "learning_rate": 0.00014,
      "loss": 2.0031,
      "step": 7
    },
    {
      "epoch": 0.0006623476083041832,
      "grad_norm": 2.190747022628784,
      "learning_rate": 0.00016,
      "loss": 1.9695,
      "step": 8
    },
    {
      "epoch": 0.000745141059342206,
      "grad_norm": 2.0315940380096436,
      "learning_rate": 0.00018,
      "loss": 1.7247,
      "step": 9
    },
    {
      "epoch": 0.000745141059342206,
      "eval_loss": 0.31743714213371277,
      "eval_runtime": 554.3092,
      "eval_samples_per_second": 9.175,
      "eval_steps_per_second": 4.588,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4353558419865600.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}