last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0003610723849834007,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 4.011915388704452e-05,
      "grad_norm": 27.923988342285156,
      "learning_rate": 2e-05,
      "loss": 10.9323,
      "step": 1
    },
    {
      "epoch": 4.011915388704452e-05,
      "eval_loss": 11.04690933227539,
      "eval_runtime": 4583.5455,
      "eval_samples_per_second": 2.29,
      "eval_steps_per_second": 1.145,
      "step": 1
    },
    {
      "epoch": 8.023830777408904e-05,
      "grad_norm": 26.460643768310547,
      "learning_rate": 4e-05,
      "loss": 10.9153,
      "step": 2
    },
    {
      "epoch": 0.00012035746166113357,
      "grad_norm": 25.659944534301758,
      "learning_rate": 6e-05,
      "loss": 11.0171,
      "step": 3
    },
    {
      "epoch": 0.00012035746166113357,
      "eval_loss": 10.575925827026367,
      "eval_runtime": 5956.9455,
      "eval_samples_per_second": 1.762,
      "eval_steps_per_second": 0.881,
      "step": 3
    },
    {
      "epoch": 0.00016047661554817808,
      "grad_norm": 27.958938598632812,
      "learning_rate": 8e-05,
      "loss": 10.4981,
      "step": 4
    },
    {
      "epoch": 0.00020059576943522262,
      "grad_norm": 38.440025329589844,
      "learning_rate": 0.0001,
      "loss": 9.6739,
      "step": 5
    },
    {
      "epoch": 0.00024071492332226714,
      "grad_norm": 33.74822998046875,
      "learning_rate": 0.00012,
      "loss": 8.801,
      "step": 6
    },
    {
      "epoch": 0.00024071492332226714,
      "eval_loss": 7.0950026512146,
      "eval_runtime": 6564.6534,
      "eval_samples_per_second": 1.599,
      "eval_steps_per_second": 0.799,
      "step": 6
    },
    {
      "epoch": 0.00028083407720931163,
      "grad_norm": 44.111427307128906,
      "learning_rate": 0.00014,
      "loss": 6.7464,
      "step": 7
    },
    {
      "epoch": 0.00032095323109635615,
      "grad_norm": 43.45552062988281,
      "learning_rate": 0.00016,
      "loss": 5.3264,
      "step": 8
    },
    {
      "epoch": 0.0003610723849834007,
      "grad_norm": 42.62737274169922,
      "learning_rate": 0.00018,
      "loss": 3.9528,
      "step": 9
    },
    {
      "epoch": 0.0003610723849834007,
      "eval_loss": 1.397772192955017,
      "eval_runtime": 2166.8639,
      "eval_samples_per_second": 4.843,
      "eval_steps_per_second": 2.422,
      "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": 648858870743040.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}