last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.04338394793926247,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004338394793926247,
      "grad_norm": 2.590311288833618,
      "learning_rate": 2e-05,
      "loss": 15.289,
      "step": 1
    },
    {
      "epoch": 0.004338394793926247,
      "eval_loss": 3.814282178878784,
      "eval_runtime": 1.7727,
      "eval_samples_per_second": 54.719,
      "eval_steps_per_second": 27.642,
      "step": 1
    },
    {
      "epoch": 0.008676789587852495,
      "grad_norm": 2.3822739124298096,
      "learning_rate": 4e-05,
      "loss": 15.2893,
      "step": 2
    },
    {
      "epoch": 0.013015184381778741,
      "grad_norm": 2.4648756980895996,
      "learning_rate": 6e-05,
      "loss": 14.282,
      "step": 3
    },
    {
      "epoch": 0.013015184381778741,
      "eval_loss": 3.8085482120513916,
      "eval_runtime": 1.7589,
      "eval_samples_per_second": 55.15,
      "eval_steps_per_second": 27.859,
      "step": 3
    },
    {
      "epoch": 0.01735357917570499,
      "grad_norm": 2.909278392791748,
      "learning_rate": 8e-05,
      "loss": 15.5635,
      "step": 4
    },
    {
      "epoch": 0.021691973969631236,
      "grad_norm": 2.349195957183838,
      "learning_rate": 0.0001,
      "loss": 15.3922,
      "step": 5
    },
    {
      "epoch": 0.026030368763557483,
      "grad_norm": 1.8822170495986938,
      "learning_rate": 0.00012,
      "loss": 15.4836,
      "step": 6
    },
    {
      "epoch": 0.026030368763557483,
      "eval_loss": 3.7893669605255127,
      "eval_runtime": 1.7797,
      "eval_samples_per_second": 54.503,
      "eval_steps_per_second": 27.532,
      "step": 6
    },
    {
      "epoch": 0.03036876355748373,
      "grad_norm": 1.790845274925232,
      "learning_rate": 0.00014,
      "loss": 14.4129,
      "step": 7
    },
    {
      "epoch": 0.03470715835140998,
      "grad_norm": 2.659196376800537,
      "learning_rate": 0.00016,
      "loss": 14.8636,
      "step": 8
    },
    {
      "epoch": 0.039045553145336226,
      "grad_norm": 2.014044761657715,
      "learning_rate": 0.00018,
      "loss": 15.2331,
      "step": 9
    },
    {
      "epoch": 0.039045553145336226,
      "eval_loss": 3.7499704360961914,
      "eval_runtime": 1.7895,
      "eval_samples_per_second": 54.205,
      "eval_steps_per_second": 27.382,
      "step": 9
    },
    {
      "epoch": 0.04338394793926247,
      "grad_norm": 1.9245425462722778,
      "learning_rate": 0.0002,
      "loss": 16.1022,
      "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": 129495151411200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}