last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0006520286240565961,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 6.520286240565961e-05,
      "grad_norm": 2.566277027130127,
      "learning_rate": 2e-05,
      "loss": 10.8809,
      "step": 1
    },
    {
      "epoch": 6.520286240565961e-05,
      "eval_loss": 2.5950257778167725,
      "eval_runtime": 345.943,
      "eval_samples_per_second": 18.668,
      "eval_steps_per_second": 9.334,
      "step": 1
    },
    {
      "epoch": 0.00013040572481131922,
      "grad_norm": 2.1301398277282715,
      "learning_rate": 4e-05,
      "loss": 10.4332,
      "step": 2
    },
    {
      "epoch": 0.00019560858721697882,
      "grad_norm": 2.9453532695770264,
      "learning_rate": 6e-05,
      "loss": 11.0066,
      "step": 3
    },
    {
      "epoch": 0.00019560858721697882,
      "eval_loss": 2.5930960178375244,
      "eval_runtime": 344.7914,
      "eval_samples_per_second": 18.73,
      "eval_steps_per_second": 9.365,
      "step": 3
    },
    {
      "epoch": 0.00026081144962263844,
      "grad_norm": 2.430074453353882,
      "learning_rate": 8e-05,
      "loss": 10.1942,
      "step": 4
    },
    {
      "epoch": 0.00032601431202829807,
      "grad_norm": 2.787076473236084,
      "learning_rate": 0.0001,
      "loss": 9.3975,
      "step": 5
    },
    {
      "epoch": 0.00039121717443395764,
      "grad_norm": 2.6258671283721924,
      "learning_rate": 0.00012,
      "loss": 10.5725,
      "step": 6
    },
    {
      "epoch": 0.00039121717443395764,
      "eval_loss": 2.5629100799560547,
      "eval_runtime": 344.0634,
      "eval_samples_per_second": 18.77,
      "eval_steps_per_second": 9.385,
      "step": 6
    },
    {
      "epoch": 0.00045642003683961726,
      "grad_norm": 2.9186737537384033,
      "learning_rate": 0.00014,
      "loss": 11.7341,
      "step": 7
    },
    {
      "epoch": 0.0005216228992452769,
      "grad_norm": 3.1749649047851562,
      "learning_rate": 0.00016,
      "loss": 9.8503,
      "step": 8
    },
    {
      "epoch": 0.0005868257616509365,
      "grad_norm": 4.3560638427734375,
      "learning_rate": 0.00018,
      "loss": 10.0232,
      "step": 9
    },
    {
      "epoch": 0.0005868257616509365,
      "eval_loss": 2.3742849826812744,
      "eval_runtime": 343.947,
      "eval_samples_per_second": 18.776,
      "eval_steps_per_second": 9.388,
      "step": 9
    },
    {
      "epoch": 0.0006520286240565961,
      "grad_norm": 4.542791843414307,
      "learning_rate": 0.0002,
      "loss": 9.6261,
      "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": 3166224305356800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}