last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.354821357992643e-05,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 9.354821357992642e-06,
      "grad_norm": 14.670015335083008,
      "learning_rate": 2e-05,
      "loss": 13.8849,
      "step": 1
    },
    {
      "epoch": 9.354821357992642e-06,
      "eval_loss": 3.4496142864227295,
      "eval_runtime": 1538.113,
      "eval_samples_per_second": 29.263,
      "eval_steps_per_second": 14.632,
      "step": 1
    },
    {
      "epoch": 1.8709642715985284e-05,
      "grad_norm": 13.07070255279541,
      "learning_rate": 4e-05,
      "loss": 14.0289,
      "step": 2
    },
    {
      "epoch": 2.8064464073977927e-05,
      "grad_norm": 14.32394027709961,
      "learning_rate": 6e-05,
      "loss": 13.3174,
      "step": 3
    },
    {
      "epoch": 2.8064464073977927e-05,
      "eval_loss": 3.3960177898406982,
      "eval_runtime": 1537.604,
      "eval_samples_per_second": 29.273,
      "eval_steps_per_second": 14.636,
      "step": 3
    },
    {
      "epoch": 3.741928543197057e-05,
      "grad_norm": 13.707303047180176,
      "learning_rate": 8e-05,
      "loss": 12.2374,
      "step": 4
    },
    {
      "epoch": 4.6774106789963214e-05,
      "grad_norm": 22.26589584350586,
      "learning_rate": 0.0001,
      "loss": 12.4981,
      "step": 5
    },
    {
      "epoch": 5.612892814795585e-05,
      "grad_norm": 12.266358375549316,
      "learning_rate": 0.00012,
      "loss": 9.7262,
      "step": 6
    },
    {
      "epoch": 5.612892814795585e-05,
      "eval_loss": 2.4447970390319824,
      "eval_runtime": 1539.323,
      "eval_samples_per_second": 29.24,
      "eval_steps_per_second": 14.62,
      "step": 6
    },
    {
      "epoch": 6.54837495059485e-05,
      "grad_norm": 19.052940368652344,
      "learning_rate": 0.00014,
      "loss": 10.6732,
      "step": 7
    },
    {
      "epoch": 7.483857086394114e-05,
      "grad_norm": 20.745851516723633,
      "learning_rate": 0.00016,
      "loss": 8.3215,
      "step": 8
    },
    {
      "epoch": 8.419339222193378e-05,
      "grad_norm": 11.37088394165039,
      "learning_rate": 0.00018,
      "loss": 5.7187,
      "step": 9
    },
    {
      "epoch": 8.419339222193378e-05,
      "eval_loss": 1.2393786907196045,
      "eval_runtime": 1539.0064,
      "eval_samples_per_second": 29.246,
      "eval_steps_per_second": 14.623,
      "step": 9
    },
    {
      "epoch": 9.354821357992643e-05,
      "grad_norm": 9.3848295211792,
      "learning_rate": 0.0002,
      "loss": 4.5529,
      "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": 1628733135912960.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}