last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.004051211468044771,
  "eval_steps": 10,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00010387721712935311,
      "eval_loss": 1.9016282558441162,
      "eval_runtime": 225.6225,
      "eval_samples_per_second": 17.968,
      "eval_steps_per_second": 8.984,
      "step": 1
    },
    {
      "epoch": 0.0005193860856467655,
      "grad_norm": 1.4621918201446533,
      "learning_rate": 5e-05,
      "loss": 1.8144,
      "step": 5
    },
    {
      "epoch": 0.001038772171293531,
      "grad_norm": 1.3856263160705566,
      "learning_rate": 0.0001,
      "loss": 1.7063,
      "step": 10
    },
    {
      "epoch": 0.001038772171293531,
      "eval_loss": 1.645436406135559,
      "eval_runtime": 225.2292,
      "eval_samples_per_second": 17.999,
      "eval_steps_per_second": 9.0,
      "step": 10
    },
    {
      "epoch": 0.0015581582569402965,
      "grad_norm": 1.6114144325256348,
      "learning_rate": 9.619397662556435e-05,
      "loss": 1.52,
      "step": 15
    },
    {
      "epoch": 0.002077544342587062,
      "grad_norm": 1.370432734489441,
      "learning_rate": 8.535533905932738e-05,
      "loss": 1.3689,
      "step": 20
    },
    {
      "epoch": 0.002077544342587062,
      "eval_loss": 1.3944305181503296,
      "eval_runtime": 224.12,
      "eval_samples_per_second": 18.089,
      "eval_steps_per_second": 9.044,
      "step": 20
    },
    {
      "epoch": 0.0025969304282338276,
      "grad_norm": 1.1585749387741089,
      "learning_rate": 6.91341716182545e-05,
      "loss": 1.3676,
      "step": 25
    },
    {
      "epoch": 0.003116316513880593,
      "grad_norm": 1.173161268234253,
      "learning_rate": 5e-05,
      "loss": 1.293,
      "step": 30
    },
    {
      "epoch": 0.003116316513880593,
      "eval_loss": 1.3442710638046265,
      "eval_runtime": 223.7354,
      "eval_samples_per_second": 18.12,
      "eval_steps_per_second": 9.06,
      "step": 30
    },
    {
      "epoch": 0.0036357025995273585,
      "grad_norm": 1.27518892288208,
      "learning_rate": 3.086582838174551e-05,
      "loss": 1.3617,
      "step": 35
    }
  ],
  "logging_steps": 5,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 13,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1987886889566208.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}