{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 11655,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3003003003003003,
      "grad_norm": 4.6720170974731445,
      "learning_rate": 1.9399399399399402e-05,
      "loss": 0.4843,
      "step": 500
    },
    {
      "epoch": 0.6006006006006006,
      "grad_norm": 17.49016571044922,
      "learning_rate": 1.87987987987988e-05,
      "loss": 0.3955,
      "step": 1000
    },
    {
      "epoch": 0.9009009009009009,
      "grad_norm": 50.96406936645508,
      "learning_rate": 1.81981981981982e-05,
      "loss": 0.3807,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.2911546230316162,
      "eval_runtime": 1.0985,
      "eval_samples_per_second": 1346.393,
      "eval_steps_per_second": 168.413,
      "step": 1665
    },
    {
      "epoch": 1.2012012012012012,
      "grad_norm": 41.78978729248047,
      "learning_rate": 1.7597597597597598e-05,
      "loss": 0.3176,
      "step": 2000
    },
    {
      "epoch": 1.5015015015015014,
      "grad_norm": 15.492409706115723,
      "learning_rate": 1.6996996996997e-05,
      "loss": 0.2729,
      "step": 2500
    },
    {
      "epoch": 1.8018018018018018,
      "grad_norm": 15.4843111038208,
      "learning_rate": 1.6396396396396396e-05,
      "loss": 0.2802,
      "step": 3000
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4967462718486786,
      "eval_runtime": 1.1211,
      "eval_samples_per_second": 1319.266,
      "eval_steps_per_second": 165.02,
      "step": 3330
    },
    {
      "epoch": 2.1021021021021022,
      "grad_norm": 0.058736126869916916,
      "learning_rate": 1.5795795795795797e-05,
      "loss": 0.2095,
      "step": 3500
    },
    {
      "epoch": 2.4024024024024024,
      "grad_norm": 41.070953369140625,
      "learning_rate": 1.5195195195195196e-05,
      "loss": 0.1309,
      "step": 4000
    },
    {
      "epoch": 2.7027027027027026,
      "grad_norm": 26.841840744018555,
      "learning_rate": 1.4594594594594596e-05,
      "loss": 0.169,
      "step": 4500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.5553244948387146,
      "eval_runtime": 1.1109,
      "eval_samples_per_second": 1331.404,
      "eval_steps_per_second": 166.538,
      "step": 4995
    },
    {
      "epoch": 3.003003003003003,
      "grad_norm": 0.2728441655635834,
      "learning_rate": 1.3993993993993995e-05,
      "loss": 0.1481,
      "step": 5000
    },
    {
      "epoch": 3.3033033033033035,
      "grad_norm": 0.013637651689350605,
      "learning_rate": 1.3393393393393394e-05,
      "loss": 0.0638,
      "step": 5500
    },
    {
      "epoch": 3.6036036036036037,
      "grad_norm": 0.5741263031959534,
      "learning_rate": 1.2792792792792795e-05,
      "loss": 0.0753,
      "step": 6000
    },
    {
      "epoch": 3.903903903903904,
      "grad_norm": 0.03777342289686203,
      "learning_rate": 1.2192192192192194e-05,
      "loss": 0.0823,
      "step": 6500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.7062426805496216,
      "eval_runtime": 1.1008,
      "eval_samples_per_second": 1343.529,
      "eval_steps_per_second": 168.055,
      "step": 6660
    },
    {
      "epoch": 4.2042042042042045,
      "grad_norm": 0.0037255873903632164,
      "learning_rate": 1.1591591591591593e-05,
      "loss": 0.0377,
      "step": 7000
    },
    {
      "epoch": 4.504504504504505,
      "grad_norm": 0.004647154361009598,
      "learning_rate": 1.0990990990990992e-05,
      "loss": 0.037,
      "step": 7500
    },
    {
      "epoch": 4.804804804804805,
      "grad_norm": 0.00761270709335804,
      "learning_rate": 1.039039039039039e-05,
      "loss": 0.0534,
      "step": 8000
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.7775081396102905,
      "eval_runtime": 1.0937,
      "eval_samples_per_second": 1352.232,
      "eval_steps_per_second": 169.143,
      "step": 8325
    },
    {
      "epoch": 5.105105105105105,
      "grad_norm": 0.004814255982637405,
      "learning_rate": 9.78978978978979e-06,
      "loss": 0.0287,
      "step": 8500
    },
    {
      "epoch": 5.405405405405405,
      "grad_norm": 0.0026015336625277996,
      "learning_rate": 9.189189189189191e-06,
      "loss": 0.025,
      "step": 9000
    },
    {
      "epoch": 5.7057057057057055,
      "grad_norm": 0.012158718891441822,
      "learning_rate": 8.588588588588589e-06,
      "loss": 0.0246,
      "step": 9500
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.8445501923561096,
      "eval_runtime": 1.0956,
      "eval_samples_per_second": 1349.885,
      "eval_steps_per_second": 168.85,
      "step": 9990
    },
    {
      "epoch": 6.006006006006006,
      "grad_norm": 0.0015228951815515757,
      "learning_rate": 7.987987987987988e-06,
      "loss": 0.0288,
      "step": 10000
    },
    {
      "epoch": 6.306306306306306,
      "grad_norm": 0.0007786727510392666,
      "learning_rate": 7.387387387387388e-06,
      "loss": 0.0152,
      "step": 10500
    },
    {
      "epoch": 6.606606606606607,
      "grad_norm": 0.0005245794309303164,
      "learning_rate": 6.786786786786788e-06,
      "loss": 0.0191,
      "step": 11000
    },
    {
      "epoch": 6.906906906906907,
      "grad_norm": 0.0009558356832712889,
      "learning_rate": 6.186186186186187e-06,
      "loss": 0.0111,
      "step": 11500
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.9164330363273621,
      "eval_runtime": 1.0989,
      "eval_samples_per_second": 1345.937,
      "eval_steps_per_second": 168.356,
      "step": 11655
    }
  ],
  "logging_steps": 500,
  "max_steps": 16650,
  "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": 3066329128047360.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}