checkpoint-300/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.7037037037037037,
  "eval_steps": 500,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.6172839506172839,
      "grad_norm": 62.88630294799805,
      "learning_rate": 9.916000000000001e-06,
      "loss": 6.6856,
      "step": 50
    },
    {
      "epoch": 1.2345679012345678,
      "grad_norm": 387.6339111328125,
      "learning_rate": 9.818000000000002e-06,
      "loss": 5.5869,
      "step": 100
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 245.18975830078125,
      "learning_rate": 9.718e-06,
      "loss": 5.159,
      "step": 150
    },
    {
      "epoch": 2.4691358024691357,
      "grad_norm": 330.3158264160156,
      "learning_rate": 9.618e-06,
      "loss": 4.0494,
      "step": 200
    },
    {
      "epoch": 3.0864197530864197,
      "grad_norm": 138.14215087890625,
      "learning_rate": 9.518000000000001e-06,
      "loss": 3.6313,
      "step": 250
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 166.69479370117188,
      "learning_rate": 9.418e-06,
      "loss": 3.3391,
      "step": 300
    }
  ],
  "logging_steps": 50,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 62,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.146729498624e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}