last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0017146776406035665,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00017146776406035664,
      "grad_norm": 59.05685043334961,
      "learning_rate": 2e-05,
      "loss": 16.22,
      "step": 1
    },
    {
      "epoch": 0.0003429355281207133,
      "grad_norm": 50.757965087890625,
      "learning_rate": 4e-05,
      "loss": 17.9129,
      "step": 2
    },
    {
      "epoch": 0.00051440329218107,
      "grad_norm": 54.17258834838867,
      "learning_rate": 6e-05,
      "loss": 15.4428,
      "step": 3
    },
    {
      "epoch": 0.00051440329218107,
      "eval_loss": 4.170095443725586,
      "eval_runtime": 54.3854,
      "eval_samples_per_second": 45.159,
      "eval_steps_per_second": 22.58,
      "step": 3
    },
    {
      "epoch": 0.0006858710562414266,
      "grad_norm": 53.12652587890625,
      "learning_rate": 8e-05,
      "loss": 17.9382,
      "step": 4
    },
    {
      "epoch": 0.0008573388203017832,
      "grad_norm": 61.29397201538086,
      "learning_rate": 0.0001,
      "loss": 15.3626,
      "step": 5
    },
    {
      "epoch": 0.00102880658436214,
      "grad_norm": 55.397151947021484,
      "learning_rate": 0.00012,
      "loss": 18.6586,
      "step": 6
    },
    {
      "epoch": 0.00102880658436214,
      "eval_loss": 4.132497787475586,
      "eval_runtime": 55.2696,
      "eval_samples_per_second": 44.437,
      "eval_steps_per_second": 22.218,
      "step": 6
    },
    {
      "epoch": 0.0012002743484224967,
      "grad_norm": 51.36023712158203,
      "learning_rate": 0.00014,
      "loss": 17.7948,
      "step": 7
    },
    {
      "epoch": 0.0013717421124828531,
      "grad_norm": 53.20388412475586,
      "learning_rate": 0.00016,
      "loss": 18.7321,
      "step": 8
    },
    {
      "epoch": 0.0015432098765432098,
      "grad_norm": 47.71489715576172,
      "learning_rate": 0.00018,
      "loss": 17.4119,
      "step": 9
    },
    {
      "epoch": 0.0015432098765432098,
      "eval_loss": 4.028396129608154,
      "eval_runtime": 55.7627,
      "eval_samples_per_second": 44.044,
      "eval_steps_per_second": 22.022,
      "step": 9
    },
    {
      "epoch": 0.0017146776406035665,
      "grad_norm": 72.37802124023438,
      "learning_rate": 0.0002,
      "loss": 16.6177,
      "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": 87728565780480.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}