{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9996261682242991,
  "eval_steps": 250,
  "global_step": 1337,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05009345794392523,
      "grad_norm": 4.856017112731934,
      "learning_rate": 4.925373134328357e-07,
      "loss": 0.6864,
      "step": 67
    },
    {
      "epoch": 0.10018691588785046,
      "grad_norm": 5.972659587860107,
      "learning_rate": 9.925373134328357e-07,
      "loss": 0.7251,
      "step": 134
    },
    {
      "epoch": 0.1502803738317757,
      "grad_norm": 3.9563512802124023,
      "learning_rate": 9.451371571072319e-07,
      "loss": 0.6276,
      "step": 201
    },
    {
      "epoch": 0.18691588785046728,
      "eval_loss": 0.530458390712738,
      "eval_runtime": 26.5207,
      "eval_samples_per_second": 18.853,
      "eval_steps_per_second": 9.427,
      "step": 250
    },
    {
      "epoch": 0.20037383177570092,
      "grad_norm": 5.4779744148254395,
      "learning_rate": 8.894430590191188e-07,
      "loss": 0.5709,
      "step": 268
    },
    {
      "epoch": 0.2504672897196262,
      "grad_norm": 3.915782928466797,
      "learning_rate": 8.337489609310058e-07,
      "loss": 0.4635,
      "step": 335
    },
    {
      "epoch": 0.3005607476635514,
      "grad_norm": 3.473511219024658,
      "learning_rate": 7.780548628428927e-07,
      "loss": 0.3408,
      "step": 402
    },
    {
      "epoch": 0.3506542056074766,
      "grad_norm": 3.5871734619140625,
      "learning_rate": 7.223607647547797e-07,
      "loss": 0.2889,
      "step": 469
    },
    {
      "epoch": 0.37383177570093457,
      "eval_loss": 0.24327994883060455,
      "eval_runtime": 26.5274,
      "eval_samples_per_second": 18.848,
      "eval_steps_per_second": 9.424,
      "step": 500
    },
    {
      "epoch": 0.40074766355140184,
      "grad_norm": 2.3577873706817627,
      "learning_rate": 6.666666666666666e-07,
      "loss": 0.2366,
      "step": 536
    },
    {
      "epoch": 0.4508411214953271,
      "grad_norm": 2.283566951751709,
      "learning_rate": 6.109725685785536e-07,
      "loss": 0.201,
      "step": 603
    },
    {
      "epoch": 0.5009345794392523,
      "grad_norm": 3.1220109462738037,
      "learning_rate": 5.552784704904405e-07,
      "loss": 0.1925,
      "step": 670
    },
    {
      "epoch": 0.5510280373831775,
      "grad_norm": 1.7564232349395752,
      "learning_rate": 4.995843724023275e-07,
      "loss": 0.1926,
      "step": 737
    },
    {
      "epoch": 0.5607476635514018,
      "eval_loss": 0.1771223098039627,
      "eval_runtime": 26.532,
      "eval_samples_per_second": 18.845,
      "eval_steps_per_second": 9.423,
      "step": 750
    },
    {
      "epoch": 0.6011214953271028,
      "grad_norm": 2.3800840377807617,
      "learning_rate": 4.438902743142144e-07,
      "loss": 0.1808,
      "step": 804
    },
    {
      "epoch": 0.6512149532710281,
      "grad_norm": 2.1542205810546875,
      "learning_rate": 3.881961762261014e-07,
      "loss": 0.1792,
      "step": 871
    },
    {
      "epoch": 0.7013084112149532,
      "grad_norm": 1.8653497695922852,
      "learning_rate": 3.3250207813798835e-07,
      "loss": 0.1655,
      "step": 938
    },
    {
      "epoch": 0.7476635514018691,
      "eval_loss": 0.15651728212833405,
      "eval_runtime": 26.5298,
      "eval_samples_per_second": 18.847,
      "eval_steps_per_second": 9.423,
      "step": 1000
    },
    {
      "epoch": 0.7514018691588785,
      "grad_norm": 1.4686954021453857,
      "learning_rate": 2.7680798004987534e-07,
      "loss": 0.1297,
      "step": 1005
    },
    {
      "epoch": 0.8014953271028037,
      "grad_norm": 1.8491053581237793,
      "learning_rate": 2.2111388196176226e-07,
      "loss": 0.1413,
      "step": 1072
    },
    {
      "epoch": 0.851588785046729,
      "grad_norm": 3.7557272911071777,
      "learning_rate": 1.6541978387364923e-07,
      "loss": 0.1543,
      "step": 1139
    },
    {
      "epoch": 0.9016822429906542,
      "grad_norm": 1.0174674987792969,
      "learning_rate": 1.0972568578553615e-07,
      "loss": 0.1353,
      "step": 1206
    },
    {
      "epoch": 0.9345794392523364,
      "eval_loss": 0.15059247612953186,
      "eval_runtime": 26.5259,
      "eval_samples_per_second": 18.85,
      "eval_steps_per_second": 9.425,
      "step": 1250
    },
    {
      "epoch": 0.9517757009345794,
      "grad_norm": 2.875340700149536,
      "learning_rate": 5.403158769742311e-08,
      "loss": 0.1429,
      "step": 1273
    },
    {
      "epoch": 0.9996261682242991,
      "step": 1337,
      "total_flos": 8.688878626214707e+16,
      "train_loss": 0.29530572784241227,
      "train_runtime": 1905.129,
      "train_samples_per_second": 5.616,
      "train_steps_per_second": 0.702
    }
  ],
  "logging_steps": 67,
  "max_steps": 1337,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "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": 8.688878626214707e+16,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}