last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.022396416573348264,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0022396416573348264,
      "grad_norm": 1.4915837049484253,
      "learning_rate": 2e-05,
      "loss": 6.0868,
      "step": 1
    },
    {
      "epoch": 0.0022396416573348264,
      "eval_loss": 5.974383354187012,
      "eval_runtime": 8.8471,
      "eval_samples_per_second": 21.25,
      "eval_steps_per_second": 10.625,
      "step": 1
    },
    {
      "epoch": 0.004479283314669653,
      "grad_norm": 1.4845587015151978,
      "learning_rate": 4e-05,
      "loss": 6.1179,
      "step": 2
    },
    {
      "epoch": 0.006718924972004479,
      "grad_norm": 1.3171576261520386,
      "learning_rate": 6e-05,
      "loss": 5.9494,
      "step": 3
    },
    {
      "epoch": 0.006718924972004479,
      "eval_loss": 5.945599555969238,
      "eval_runtime": 8.8447,
      "eval_samples_per_second": 21.256,
      "eval_steps_per_second": 10.628,
      "step": 3
    },
    {
      "epoch": 0.008958566629339306,
      "grad_norm": 1.0633552074432373,
      "learning_rate": 8e-05,
      "loss": 5.781,
      "step": 4
    },
    {
      "epoch": 0.011198208286674132,
      "grad_norm": 1.2751410007476807,
      "learning_rate": 0.0001,
      "loss": 5.8456,
      "step": 5
    },
    {
      "epoch": 0.013437849944008958,
      "grad_norm": 0.9956833124160767,
      "learning_rate": 0.00012,
      "loss": 5.8907,
      "step": 6
    },
    {
      "epoch": 0.013437849944008958,
      "eval_loss": 5.812419414520264,
      "eval_runtime": 8.9729,
      "eval_samples_per_second": 20.952,
      "eval_steps_per_second": 10.476,
      "step": 6
    },
    {
      "epoch": 0.015677491601343786,
      "grad_norm": 1.2121307849884033,
      "learning_rate": 0.00014,
      "loss": 5.9106,
      "step": 7
    },
    {
      "epoch": 0.01791713325867861,
      "grad_norm": 1.2964277267456055,
      "learning_rate": 0.00016,
      "loss": 5.8155,
      "step": 8
    },
    {
      "epoch": 0.020156774916013438,
      "grad_norm": 1.2249387502670288,
      "learning_rate": 0.00018,
      "loss": 5.7872,
      "step": 9
    },
    {
      "epoch": 0.020156774916013438,
      "eval_loss": 5.777248859405518,
      "eval_runtime": 8.99,
      "eval_samples_per_second": 20.912,
      "eval_steps_per_second": 10.456,
      "step": 9
    },
    {
      "epoch": 0.022396416573348264,
      "grad_norm": 0.9850199222564697,
      "learning_rate": 0.0002,
      "loss": 5.7362,
      "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": 2052393617326080.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}