last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.006578947368421052,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006578947368421052,
      "grad_norm": 0.29057952761650085,
      "learning_rate": 2e-05,
      "loss": 2.2014,
      "step": 1
    },
    {
      "epoch": 0.0006578947368421052,
      "eval_loss": 2.1872005462646484,
      "eval_runtime": 32.0967,
      "eval_samples_per_second": 19.94,
      "eval_steps_per_second": 9.97,
      "step": 1
    },
    {
      "epoch": 0.0013157894736842105,
      "grad_norm": 0.3000072240829468,
      "learning_rate": 4e-05,
      "loss": 2.0864,
      "step": 2
    },
    {
      "epoch": 0.001973684210526316,
      "grad_norm": 0.3774920701980591,
      "learning_rate": 6e-05,
      "loss": 1.8095,
      "step": 3
    },
    {
      "epoch": 0.001973684210526316,
      "eval_loss": 2.18607234954834,
      "eval_runtime": 30.6633,
      "eval_samples_per_second": 20.872,
      "eval_steps_per_second": 10.436,
      "step": 3
    },
    {
      "epoch": 0.002631578947368421,
      "grad_norm": 0.2742858827114105,
      "learning_rate": 8e-05,
      "loss": 2.4152,
      "step": 4
    },
    {
      "epoch": 0.003289473684210526,
      "grad_norm": 0.47108742594718933,
      "learning_rate": 0.0001,
      "loss": 2.4476,
      "step": 5
    },
    {
      "epoch": 0.003947368421052632,
      "grad_norm": 0.3023812472820282,
      "learning_rate": 0.00012,
      "loss": 2.2412,
      "step": 6
    },
    {
      "epoch": 0.003947368421052632,
      "eval_loss": 2.1723532676696777,
      "eval_runtime": 30.8929,
      "eval_samples_per_second": 20.717,
      "eval_steps_per_second": 10.358,
      "step": 6
    },
    {
      "epoch": 0.004605263157894736,
      "grad_norm": 0.3247911334037781,
      "learning_rate": 0.00014,
      "loss": 1.9942,
      "step": 7
    },
    {
      "epoch": 0.005263157894736842,
      "grad_norm": 0.4075062572956085,
      "learning_rate": 0.00016,
      "loss": 2.4667,
      "step": 8
    },
    {
      "epoch": 0.0059210526315789476,
      "grad_norm": 0.507950484752655,
      "learning_rate": 0.00018,
      "loss": 2.1074,
      "step": 9
    },
    {
      "epoch": 0.0059210526315789476,
      "eval_loss": 2.1078224182128906,
      "eval_runtime": 30.8858,
      "eval_samples_per_second": 20.721,
      "eval_steps_per_second": 10.361,
      "step": 9
    },
    {
      "epoch": 0.006578947368421052,
      "grad_norm": 0.4082265794277191,
      "learning_rate": 0.0002,
      "loss": 2.2388,
      "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": 2801298969722880.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}