{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8738394320043692,
  "eval_steps": 500,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10922992900054615,
      "grad_norm": 0.9561710357666016,
      "learning_rate": 1.4564520827264783e-05,
      "loss": 0.7634,
      "step": 500
    },
    {
      "epoch": 0.10922992900054615,
      "eval_loss": 0.3866455852985382,
      "eval_runtime": 53.8542,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 500
    },
    {
      "epoch": 0.2184598580010923,
      "grad_norm": 0.6381384134292603,
      "learning_rate": 2.9129041654529566e-05,
      "loss": 0.2265,
      "step": 1000
    },
    {
      "epoch": 0.2184598580010923,
      "eval_loss": 0.28045815229415894,
      "eval_runtime": 53.8781,
      "eval_samples_per_second": 1.726,
      "eval_steps_per_second": 1.726,
      "step": 1000
    },
    {
      "epoch": 0.3276897870016384,
      "grad_norm": 0.80877685546875,
      "learning_rate": 4.3693562481794354e-05,
      "loss": 0.13,
      "step": 1500
    },
    {
      "epoch": 0.3276897870016384,
      "eval_loss": 0.24824748933315277,
      "eval_runtime": 53.7906,
      "eval_samples_per_second": 1.729,
      "eval_steps_per_second": 1.729,
      "step": 1500
    },
    {
      "epoch": 0.4369197160021846,
      "grad_norm": 0.7288264632225037,
      "learning_rate": 5.8228954267404603e-05,
      "loss": 0.1067,
      "step": 2000
    },
    {
      "epoch": 0.4369197160021846,
      "eval_loss": 0.24914385378360748,
      "eval_runtime": 53.8469,
      "eval_samples_per_second": 1.727,
      "eval_steps_per_second": 1.727,
      "step": 2000
    },
    {
      "epoch": 0.5461496450027308,
      "grad_norm": 1.1100046634674072,
      "learning_rate": 7.27934750946694e-05,
      "loss": 0.1018,
      "step": 2500
    },
    {
      "epoch": 0.5461496450027308,
      "eval_loss": 0.27210181951522827,
      "eval_runtime": 53.5513,
      "eval_samples_per_second": 1.737,
      "eval_steps_per_second": 1.737,
      "step": 2500
    },
    {
      "epoch": 0.6553795740032768,
      "grad_norm": 0.8480390310287476,
      "learning_rate": 8.732886688027965e-05,
      "loss": 0.0976,
      "step": 3000
    },
    {
      "epoch": 0.6553795740032768,
      "eval_loss": 0.2787032425403595,
      "eval_runtime": 51.6926,
      "eval_samples_per_second": 1.799,
      "eval_steps_per_second": 1.799,
      "step": 3000
    },
    {
      "epoch": 0.764609503003823,
      "grad_norm": 0.8974778056144714,
      "learning_rate": 0.00010189338770754442,
      "loss": 0.0948,
      "step": 3500
    },
    {
      "epoch": 0.764609503003823,
      "eval_loss": 0.26553070545196533,
      "eval_runtime": 52.0101,
      "eval_samples_per_second": 1.788,
      "eval_steps_per_second": 1.788,
      "step": 3500
    },
    {
      "epoch": 0.8738394320043692,
      "grad_norm": 0.4318696856498718,
      "learning_rate": 0.00011645790853480921,
      "loss": 0.0941,
      "step": 4000
    },
    {
      "epoch": 0.8738394320043692,
      "eval_loss": 0.2772391140460968,
      "eval_runtime": 51.821,
      "eval_samples_per_second": 1.795,
      "eval_steps_per_second": 1.795,
      "step": 4000
    }
  ],
  "logging_steps": 500,
  "max_steps": 228850,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 8.325632006613197e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}