trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9743589743589745,
  "eval_steps": 500,
  "global_step": 87,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3418803418803419,
      "grad_norm": 26.595144719965496,
      "learning_rate": 5e-06,
      "loss": 1.1527,
      "step": 10
    },
    {
      "epoch": 0.6837606837606838,
      "grad_norm": 4.417681633599547,
      "learning_rate": 5e-06,
      "loss": 1.0586,
      "step": 20
    },
    {
      "epoch": 0.9914529914529915,
      "eval_loss": 0.9911791086196899,
      "eval_runtime": 21.2936,
      "eval_samples_per_second": 37.006,
      "eval_steps_per_second": 0.611,
      "step": 29
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 6.947024802510117,
      "learning_rate": 5e-06,
      "loss": 1.0822,
      "step": 30
    },
    {
      "epoch": 1.3675213675213675,
      "grad_norm": 1.1321879610756762,
      "learning_rate": 5e-06,
      "loss": 0.9553,
      "step": 40
    },
    {
      "epoch": 1.7094017094017095,
      "grad_norm": 1.3295902454206836,
      "learning_rate": 5e-06,
      "loss": 0.9318,
      "step": 50
    },
    {
      "epoch": 1.982905982905983,
      "eval_loss": 0.9310455918312073,
      "eval_runtime": 20.72,
      "eval_samples_per_second": 38.031,
      "eval_steps_per_second": 0.627,
      "step": 58
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 1.0685677473093818,
      "learning_rate": 5e-06,
      "loss": 0.9805,
      "step": 60
    },
    {
      "epoch": 2.393162393162393,
      "grad_norm": 1.2586309486541096,
      "learning_rate": 5e-06,
      "loss": 0.8658,
      "step": 70
    },
    {
      "epoch": 2.735042735042735,
      "grad_norm": 1.1566330504038362,
      "learning_rate": 5e-06,
      "loss": 0.8547,
      "step": 80
    },
    {
      "epoch": 2.9743589743589745,
      "eval_loss": 0.9065744876861572,
      "eval_runtime": 19.5223,
      "eval_samples_per_second": 40.364,
      "eval_steps_per_second": 0.666,
      "step": 87
    },
    {
      "epoch": 2.9743589743589745,
      "step": 87,
      "total_flos": 145518860697600.0,
      "train_loss": 0.9746759622946553,
      "train_runtime": 3122.9131,
      "train_samples_per_second": 14.367,
      "train_steps_per_second": 0.028
    }
  ],
  "logging_steps": 10,
  "max_steps": 87,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 145518860697600.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}