{
  "best_metric": 0.044874656945466995,
  "best_model_checkpoint": "./results/checkpoint-8788",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 8788,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1137915339098771,
      "grad_norm": 29.75603675842285,
      "learning_rate": 4.9431042330450616e-05,
      "loss": 0.364,
      "step": 500
    },
    {
      "epoch": 0.2275830678197542,
      "grad_norm": 0.03848416730761528,
      "learning_rate": 4.886208466090123e-05,
      "loss": 0.1224,
      "step": 1000
    },
    {
      "epoch": 0.3413746017296313,
      "grad_norm": 0.22520601749420166,
      "learning_rate": 4.829312699135184e-05,
      "loss": 0.096,
      "step": 1500
    },
    {
      "epoch": 0.4551661356395084,
      "grad_norm": 46.81020736694336,
      "learning_rate": 4.772416932180246e-05,
      "loss": 0.0691,
      "step": 2000
    },
    {
      "epoch": 0.5689576695493855,
      "grad_norm": 37.5096321105957,
      "learning_rate": 4.7155211652253075e-05,
      "loss": 0.073,
      "step": 2500
    },
    {
      "epoch": 0.6827492034592626,
      "grad_norm": 0.003177793463692069,
      "learning_rate": 4.658625398270369e-05,
      "loss": 0.0574,
      "step": 3000
    },
    {
      "epoch": 0.7965407373691398,
      "grad_norm": 27.808433532714844,
      "learning_rate": 4.60172963131543e-05,
      "loss": 0.0706,
      "step": 3500
    },
    {
      "epoch": 0.9103322712790168,
      "grad_norm": 0.015329302288591862,
      "learning_rate": 4.5448338643604914e-05,
      "loss": 0.0539,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.0667862519621849,
      "eval_runtime": 508.789,
      "eval_samples_per_second": 34.537,
      "eval_steps_per_second": 2.16,
      "step": 4394
    },
    {
      "epoch": 1.024123805188894,
      "grad_norm": 0.0013099159114062786,
      "learning_rate": 4.4879380974055534e-05,
      "loss": 0.0558,
      "step": 4500
    },
    {
      "epoch": 1.137915339098771,
      "grad_norm": 3.2377407550811768,
      "learning_rate": 4.431042330450615e-05,
      "loss": 0.0247,
      "step": 5000
    },
    {
      "epoch": 1.251706873008648,
      "grad_norm": 0.05278199911117554,
      "learning_rate": 4.374146563495676e-05,
      "loss": 0.0375,
      "step": 5500
    },
    {
      "epoch": 1.3654984069185252,
      "grad_norm": 0.0037405244074761868,
      "learning_rate": 4.317250796540738e-05,
      "loss": 0.0619,
      "step": 6000
    },
    {
      "epoch": 1.4792899408284024,
      "grad_norm": 1.2426213026046753,
      "learning_rate": 4.260355029585799e-05,
      "loss": 0.0431,
      "step": 6500
    },
    {
      "epoch": 1.5930814747382795,
      "grad_norm": 0.1503586322069168,
      "learning_rate": 4.2034592626308606e-05,
      "loss": 0.0285,
      "step": 7000
    },
    {
      "epoch": 1.7068730086481567,
      "grad_norm": 0.00018984633788932115,
      "learning_rate": 4.146563495675922e-05,
      "loss": 0.0354,
      "step": 7500
    },
    {
      "epoch": 1.8206645425580337,
      "grad_norm": 0.13596081733703613,
      "learning_rate": 4.089667728720983e-05,
      "loss": 0.042,
      "step": 8000
    },
    {
      "epoch": 1.9344560764679106,
      "grad_norm": 0.37792855501174927,
      "learning_rate": 4.0327719617660446e-05,
      "loss": 0.0337,
      "step": 8500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.044874656945466995,
      "eval_runtime": 509.6348,
      "eval_samples_per_second": 34.48,
      "eval_steps_per_second": 2.156,
      "step": 8788
    }
  ],
  "logging_steps": 500,
  "max_steps": 43940,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.821569115965866e+19,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}