last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.47619047619047616,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.047619047619047616,
      "grad_norm": 3.2943832874298096,
      "learning_rate": 2e-05,
      "loss": 3.6344,
      "step": 1
    },
    {
      "epoch": 0.047619047619047616,
      "eval_loss": 4.260468006134033,
      "eval_runtime": 1.7086,
      "eval_samples_per_second": 5.268,
      "eval_steps_per_second": 2.926,
      "step": 1
    },
    {
      "epoch": 0.09523809523809523,
      "grad_norm": 3.340632915496826,
      "learning_rate": 4e-05,
      "loss": 3.1556,
      "step": 2
    },
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 3.4076058864593506,
      "learning_rate": 6e-05,
      "loss": 4.9251,
      "step": 3
    },
    {
      "epoch": 0.14285714285714285,
      "eval_loss": 4.248776435852051,
      "eval_runtime": 0.3316,
      "eval_samples_per_second": 27.143,
      "eval_steps_per_second": 15.08,
      "step": 3
    },
    {
      "epoch": 0.19047619047619047,
      "grad_norm": 4.312319755554199,
      "learning_rate": 8e-05,
      "loss": 4.1059,
      "step": 4
    },
    {
      "epoch": 0.23809523809523808,
      "grad_norm": 3.631556987762451,
      "learning_rate": 0.0001,
      "loss": 4.1648,
      "step": 5
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 4.5404767990112305,
      "learning_rate": 0.00012,
      "loss": 4.7157,
      "step": 6
    },
    {
      "epoch": 0.2857142857142857,
      "eval_loss": 3.9749979972839355,
      "eval_runtime": 0.332,
      "eval_samples_per_second": 27.11,
      "eval_steps_per_second": 15.061,
      "step": 6
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 4.91957950592041,
      "learning_rate": 0.00014,
      "loss": 4.252,
      "step": 7
    },
    {
      "epoch": 0.38095238095238093,
      "grad_norm": 6.185959339141846,
      "learning_rate": 0.00016,
      "loss": 4.3717,
      "step": 8
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 4.06405782699585,
      "learning_rate": 0.00018,
      "loss": 3.3945,
      "step": 9
    },
    {
      "epoch": 0.42857142857142855,
      "eval_loss": 3.1844916343688965,
      "eval_runtime": 0.3315,
      "eval_samples_per_second": 27.153,
      "eval_steps_per_second": 15.085,
      "step": 9
    },
    {
      "epoch": 0.47619047619047616,
      "grad_norm": 4.348942756652832,
      "learning_rate": 0.0002,
      "loss": 3.3159,
      "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": 1849564248145920.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}