last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00027525839882189403,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.7525839882189405e-05,
      "grad_norm": 0.21146489679813385,
      "learning_rate": 2e-05,
      "loss": 0.2077,
      "step": 1
    },
    {
      "epoch": 2.7525839882189405e-05,
      "eval_loss": 0.20811736583709717,
      "eval_runtime": 832.0425,
      "eval_samples_per_second": 18.385,
      "eval_steps_per_second": 9.193,
      "step": 1
    },
    {
      "epoch": 5.505167976437881e-05,
      "grad_norm": 0.20771676301956177,
      "learning_rate": 4e-05,
      "loss": 0.1816,
      "step": 2
    },
    {
      "epoch": 8.257751964656821e-05,
      "grad_norm": 0.2978658676147461,
      "learning_rate": 6e-05,
      "loss": 0.2066,
      "step": 3
    },
    {
      "epoch": 8.257751964656821e-05,
      "eval_loss": 0.2068672478199005,
      "eval_runtime": 830.7761,
      "eval_samples_per_second": 18.413,
      "eval_steps_per_second": 9.207,
      "step": 3
    },
    {
      "epoch": 0.00011010335952875762,
      "grad_norm": 0.18490785360336304,
      "learning_rate": 8e-05,
      "loss": 0.1587,
      "step": 4
    },
    {
      "epoch": 0.00013762919941094702,
      "grad_norm": 0.6858919858932495,
      "learning_rate": 0.0001,
      "loss": 0.2608,
      "step": 5
    },
    {
      "epoch": 0.00016515503929313643,
      "grad_norm": 0.2280646711587906,
      "learning_rate": 0.00012,
      "loss": 0.1434,
      "step": 6
    },
    {
      "epoch": 0.00016515503929313643,
      "eval_loss": 0.18119999766349792,
      "eval_runtime": 830.9926,
      "eval_samples_per_second": 18.408,
      "eval_steps_per_second": 9.205,
      "step": 6
    },
    {
      "epoch": 0.00019268087917532583,
      "grad_norm": 0.29051458835601807,
      "learning_rate": 0.00014,
      "loss": 0.1689,
      "step": 7
    },
    {
      "epoch": 0.00022020671905751524,
      "grad_norm": 0.3273032605648041,
      "learning_rate": 0.00016,
      "loss": 0.161,
      "step": 8
    },
    {
      "epoch": 0.00024773255893970465,
      "grad_norm": 0.3765033483505249,
      "learning_rate": 0.00018,
      "loss": 0.1222,
      "step": 9
    },
    {
      "epoch": 0.00024773255893970465,
      "eval_loss": 0.10314656049013138,
      "eval_runtime": 831.0752,
      "eval_samples_per_second": 18.406,
      "eval_steps_per_second": 9.204,
      "step": 9
    },
    {
      "epoch": 0.00027525839882189403,
      "grad_norm": 0.17171159386634827,
      "learning_rate": 0.0002,
      "loss": 0.073,
      "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": 3329215646662656.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}