trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9388753056234719,
  "eval_steps": 500,
  "global_step": 12,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07823960880195599,
      "grad_norm": 339.19366455078125,
      "learning_rate": 1.5e-06,
      "loss": 1.7327,
      "step": 1
    },
    {
      "epoch": 0.15647921760391198,
      "grad_norm": 343.84027099609375,
      "learning_rate": 3e-06,
      "loss": 1.6942,
      "step": 2
    },
    {
      "epoch": 0.23471882640586797,
      "grad_norm": 55.34988784790039,
      "learning_rate": 4.5e-06,
      "loss": 0.2441,
      "step": 3
    },
    {
      "epoch": 0.31295843520782396,
      "grad_norm": 22.917522430419922,
      "learning_rate": 6e-06,
      "loss": 0.0951,
      "step": 4
    },
    {
      "epoch": 0.39119804400978,
      "grad_norm": 4.5077643394470215,
      "learning_rate": 7.5e-06,
      "loss": 0.0653,
      "step": 5
    },
    {
      "epoch": 0.46943765281173594,
      "grad_norm": 16.18638801574707,
      "learning_rate": 9e-06,
      "loss": 0.1845,
      "step": 6
    },
    {
      "epoch": 0.5476772616136919,
      "grad_norm": 3.779344081878662,
      "learning_rate": 1.05e-05,
      "loss": 0.0566,
      "step": 7
    },
    {
      "epoch": 0.6259168704156479,
      "grad_norm": 2.3550302982330322,
      "learning_rate": 1.2e-05,
      "loss": 0.0463,
      "step": 8
    },
    {
      "epoch": 0.7041564792176039,
      "grad_norm": 3.7851593494415283,
      "learning_rate": 1.3500000000000001e-05,
      "loss": 0.0531,
      "step": 9
    },
    {
      "epoch": 0.78239608801956,
      "grad_norm": 5.353140354156494,
      "learning_rate": 1.5e-05,
      "loss": 0.0633,
      "step": 10
    },
    {
      "epoch": 0.8606356968215159,
      "grad_norm": 3.454190969467163,
      "learning_rate": 1.65e-05,
      "loss": 0.0633,
      "step": 11
    },
    {
      "epoch": 0.9388753056234719,
      "grad_norm": 8.502099990844727,
      "learning_rate": 1.8e-05,
      "loss": 0.0409,
      "step": 12
    },
    {
      "epoch": 0.9388753056234719,
      "eval_loss": 0.041706331074237823,
      "eval_runtime": 69.2407,
      "eval_samples_per_second": 1.314,
      "eval_steps_per_second": 1.314,
      "step": 12
    }
  ],
  "logging_steps": 1,
  "max_steps": 36,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 2.4308284688464896e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}