trainer_state.json

{
  "best_metric": 1.7112983465194702,
  "best_model_checkpoint": "out/checkpoint-5",
  "epoch": 0.008531383302864107,
  "eval_steps": 1,
  "global_step": 7,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 2.2998228073120117,
      "learning_rate": 0.0002,
      "loss": 2.2838,
      "step": 1
    },
    {
      "epoch": 0.0,
      "eval_loss": 2.0825579166412354,
      "eval_runtime": 2259.3163,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 2.200953245162964,
      "learning_rate": 0.0002,
      "loss": 2.059,
      "step": 2
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.9111530780792236,
      "eval_runtime": 2259.4545,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.8299978971481323,
      "learning_rate": 0.0002,
      "loss": 1.9217,
      "step": 3
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.8130964040756226,
      "eval_runtime": 2259.5777,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 3
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.5078331232070923,
      "learning_rate": 0.0002,
      "loss": 1.6955,
      "step": 4
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.7504655122756958,
      "eval_runtime": 2259.3668,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 4
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.3721721172332764,
      "learning_rate": 0.0002,
      "loss": 1.7828,
      "step": 5
    },
    {
      "epoch": 0.01,
      "eval_loss": 1.7112983465194702,
      "eval_runtime": 2259.3548,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": NaN,
      "learning_rate": 0.0002,
      "loss": 2.0711,
      "step": 6
    },
    {
      "epoch": 0.01,
      "eval_loss": 1.7112983465194702,
      "eval_runtime": 2259.3294,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 6
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.5560696125030518,
      "learning_rate": 0.0002,
      "loss": 1.713,
      "step": 7
    },
    {
      "epoch": 0.01,
      "eval_loss": 1.7407238483428955,
      "eval_runtime": 2259.3516,
      "eval_samples_per_second": 0.366,
      "eval_steps_per_second": 0.366,
      "step": 7
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1,
  "total_flos": 1226868850163712.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}