last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00039878370968545937,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.9878370968545935e-05,
      "grad_norm": 28.375574111938477,
      "learning_rate": 2e-05,
      "loss": 10.1211,
      "step": 1
    },
    {
      "epoch": 3.9878370968545935e-05,
      "eval_loss": 10.486047744750977,
      "eval_runtime": 461.401,
      "eval_samples_per_second": 22.885,
      "eval_steps_per_second": 11.443,
      "step": 1
    },
    {
      "epoch": 7.975674193709187e-05,
      "grad_norm": 46.455833435058594,
      "learning_rate": 4e-05,
      "loss": 10.324,
      "step": 2
    },
    {
      "epoch": 0.0001196351129056378,
      "grad_norm": 31.14690399169922,
      "learning_rate": 6e-05,
      "loss": 9.5556,
      "step": 3
    },
    {
      "epoch": 0.0001196351129056378,
      "eval_loss": 9.934382438659668,
      "eval_runtime": 460.1872,
      "eval_samples_per_second": 22.945,
      "eval_steps_per_second": 11.474,
      "step": 3
    },
    {
      "epoch": 0.00015951348387418374,
      "grad_norm": 35.35398483276367,
      "learning_rate": 8e-05,
      "loss": 10.4589,
      "step": 4
    },
    {
      "epoch": 0.00019939185484272969,
      "grad_norm": 37.92613220214844,
      "learning_rate": 0.0001,
      "loss": 7.7222,
      "step": 5
    },
    {
      "epoch": 0.0002392702258112756,
      "grad_norm": 36.25010299682617,
      "learning_rate": 0.00012,
      "loss": 4.6559,
      "step": 6
    },
    {
      "epoch": 0.0002392702258112756,
      "eval_loss": 3.3957276344299316,
      "eval_runtime": 460.4612,
      "eval_samples_per_second": 22.931,
      "eval_steps_per_second": 11.467,
      "step": 6
    },
    {
      "epoch": 0.00027914859677982153,
      "grad_norm": 23.317548751831055,
      "learning_rate": 0.00014,
      "loss": 5.6145,
      "step": 7
    },
    {
      "epoch": 0.0003190269677483675,
      "grad_norm": 10.3938627243042,
      "learning_rate": 0.00016,
      "loss": 1.624,
      "step": 8
    },
    {
      "epoch": 0.0003589053387169134,
      "grad_norm": 20.77987289428711,
      "learning_rate": 0.00018,
      "loss": 2.2682,
      "step": 9
    },
    {
      "epoch": 0.0003589053387169134,
      "eval_loss": 1.820220947265625,
      "eval_runtime": 460.2945,
      "eval_samples_per_second": 22.94,
      "eval_steps_per_second": 11.471,
      "step": 9
    },
    {
      "epoch": 0.00039878370968545937,
      "grad_norm": 16.127281188964844,
      "learning_rate": 0.0002,
      "loss": 3.3854,
      "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": 2820585478422528.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}