last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00011256539345826216,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.1256539345826217e-05,
      "grad_norm": 4.918504238128662,
      "learning_rate": 2e-05,
      "loss": 17.1448,
      "step": 1
    },
    {
      "epoch": 1.1256539345826217e-05,
      "eval_loss": 4.492743015289307,
      "eval_runtime": 2072.2543,
      "eval_samples_per_second": 18.051,
      "eval_steps_per_second": 9.025,
      "step": 1
    },
    {
      "epoch": 2.2513078691652433e-05,
      "grad_norm": 11.243107795715332,
      "learning_rate": 4e-05,
      "loss": 19.7466,
      "step": 2
    },
    {
      "epoch": 3.3769618037478645e-05,
      "grad_norm": 5.093787670135498,
      "learning_rate": 6e-05,
      "loss": 18.6519,
      "step": 3
    },
    {
      "epoch": 3.3769618037478645e-05,
      "eval_loss": 4.478453636169434,
      "eval_runtime": 2068.4701,
      "eval_samples_per_second": 18.084,
      "eval_steps_per_second": 9.042,
      "step": 3
    },
    {
      "epoch": 4.502615738330487e-05,
      "grad_norm": 8.79734992980957,
      "learning_rate": 8e-05,
      "loss": 17.3248,
      "step": 4
    },
    {
      "epoch": 5.628269672913108e-05,
      "grad_norm": 6.966519355773926,
      "learning_rate": 0.0001,
      "loss": 18.4569,
      "step": 5
    },
    {
      "epoch": 6.753923607495729e-05,
      "grad_norm": 7.464043140411377,
      "learning_rate": 0.00012,
      "loss": 17.1601,
      "step": 6
    },
    {
      "epoch": 6.753923607495729e-05,
      "eval_loss": 4.197001934051514,
      "eval_runtime": 2063.2511,
      "eval_samples_per_second": 18.13,
      "eval_steps_per_second": 9.065,
      "step": 6
    },
    {
      "epoch": 7.879577542078351e-05,
      "grad_norm": 6.941649913787842,
      "learning_rate": 0.00014,
      "loss": 20.4721,
      "step": 7
    },
    {
      "epoch": 9.005231476660973e-05,
      "grad_norm": 8.234708786010742,
      "learning_rate": 0.00016,
      "loss": 14.1716,
      "step": 8
    },
    {
      "epoch": 0.00010130885411243594,
      "grad_norm": 10.47228717803955,
      "learning_rate": 0.00018,
      "loss": 17.4619,
      "step": 9
    },
    {
      "epoch": 0.00010130885411243594,
      "eval_loss": 3.2568111419677734,
      "eval_runtime": 2066.8912,
      "eval_samples_per_second": 18.098,
      "eval_steps_per_second": 9.049,
      "step": 9
    },
    {
      "epoch": 0.00011256539345826216,
      "grad_norm": 11.171597480773926,
      "learning_rate": 0.0002,
      "loss": 13.1229,
      "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": 3166224305356800.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}