{
  "best_metric": 1.0519311428070068,
  "best_model_checkpoint": "miner_id_24/checkpoint-10",
  "epoch": 0.044296788482835,
  "eval_steps": 5,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004429678848283499,
      "eval_loss": 1.2574131488800049,
      "eval_runtime": 26.372,
      "eval_samples_per_second": 7.205,
      "eval_steps_per_second": 1.82,
      "step": 1
    },
    {
      "epoch": 0.013289036544850499,
      "grad_norm": 4.626389980316162,
      "learning_rate": 0.00012,
      "loss": 4.871,
      "step": 3
    },
    {
      "epoch": 0.0221483942414175,
      "eval_loss": 1.1578865051269531,
      "eval_runtime": 26.8269,
      "eval_samples_per_second": 7.082,
      "eval_steps_per_second": 1.789,
      "step": 5
    },
    {
      "epoch": 0.026578073089700997,
      "grad_norm": 3.5244829654693604,
      "learning_rate": 0.0001992114701314478,
      "loss": 4.5071,
      "step": 6
    },
    {
      "epoch": 0.03986710963455149,
      "grad_norm": 3.0460095405578613,
      "learning_rate": 0.00018763066800438636,
      "loss": 4.0276,
      "step": 9
    },
    {
      "epoch": 0.044296788482835,
      "eval_loss": 1.0519311428070068,
      "eval_runtime": 26.8819,
      "eval_samples_per_second": 7.068,
      "eval_steps_per_second": 1.786,
      "step": 10
    }
  ],
  "logging_steps": 3,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 1,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 7031294986813440.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}