{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 100,
  "global_step": 169,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.029585798816568046,
      "grad_norm": 2.40625,
      "learning_rate": 5.882352941176471e-06,
      "loss": 1.1011,
      "mean_token_accuracy": 0.7099674616006506,
      "step": 5
    },
    {
      "epoch": 0.05917159763313609,
      "grad_norm": 1.46875,
      "learning_rate": 1.1764705882352942e-05,
      "loss": 1.0735,
      "mean_token_accuracy": 0.7143687536382451,
      "step": 10
    },
    {
      "epoch": 0.08875739644970414,
      "grad_norm": 0.93359375,
      "learning_rate": 1.7647058823529414e-05,
      "loss": 1.0246,
      "mean_token_accuracy": 0.7213007797302202,
      "step": 15
    },
    {
      "epoch": 0.11834319526627218,
      "grad_norm": 0.61328125,
      "learning_rate": 1.9980782984658682e-05,
      "loss": 0.9457,
      "mean_token_accuracy": 0.7365057897688324,
      "step": 20
    },
    {
      "epoch": 0.14792899408284024,
      "grad_norm": 0.51171875,
      "learning_rate": 1.9863613034027224e-05,
      "loss": 0.9101,
      "mean_token_accuracy": 0.7418338168299983,
      "step": 25
    },
    {
      "epoch": 0.17751479289940827,
      "grad_norm": 0.42578125,
      "learning_rate": 1.9641197940012136e-05,
      "loss": 0.8957,
      "mean_token_accuracy": 0.7444947984183858,
      "step": 30
    },
    {
      "epoch": 0.20710059171597633,
      "grad_norm": 0.408203125,
      "learning_rate": 1.9315910880512792e-05,
      "loss": 0.8774,
      "mean_token_accuracy": 0.7473870896289159,
      "step": 35
    },
    {
      "epoch": 0.23668639053254437,
      "grad_norm": 0.390625,
      "learning_rate": 1.8891222681391853e-05,
      "loss": 0.8628,
      "mean_token_accuracy": 0.7510183565770613,
      "step": 40
    },
    {
      "epoch": 0.26627218934911245,
      "grad_norm": 0.35546875,
      "learning_rate": 1.8371664782625287e-05,
      "loss": 0.8562,
      "mean_token_accuracy": 0.7520073491134827,
      "step": 45
    },
    {
      "epoch": 0.2958579881656805,
      "grad_norm": 0.32421875,
      "learning_rate": 1.7762780887657576e-05,
      "loss": 0.8523,
      "mean_token_accuracy": 0.7526434348741664,
      "step": 50
    },
    {
      "epoch": 0.3254437869822485,
      "grad_norm": 0.341796875,
      "learning_rate": 1.7071067811865477e-05,
      "loss": 0.8499,
      "mean_token_accuracy": 0.7530957012263093,
      "step": 55
    },
    {
      "epoch": 0.35502958579881655,
      "grad_norm": 0.34765625,
      "learning_rate": 1.6303906161279554e-05,
      "loss": 0.862,
      "mean_token_accuracy": 0.7495981366725619,
      "step": 60
    },
    {
      "epoch": 0.38461538461538464,
      "grad_norm": 0.32421875,
      "learning_rate": 1.5469481581224274e-05,
      "loss": 0.8489,
      "mean_token_accuracy": 0.7524725022168722,
      "step": 65
    },
    {
      "epoch": 0.41420118343195267,
      "grad_norm": 0.337890625,
      "learning_rate": 1.4576697415156818e-05,
      "loss": 0.8416,
      "mean_token_accuracy": 0.7547405747257215,
      "step": 70
    },
    {
      "epoch": 0.4437869822485207,
      "grad_norm": 0.3125,
      "learning_rate": 1.3635079705638298e-05,
      "loss": 0.8415,
      "mean_token_accuracy": 0.7544626406760742,
      "step": 75
    },
    {
      "epoch": 0.47337278106508873,
      "grad_norm": 0.322265625,
      "learning_rate": 1.2654675551080724e-05,
      "loss": 0.8448,
      "mean_token_accuracy": 0.7532455186182244,
      "step": 80
    },
    {
      "epoch": 0.5029585798816568,
      "grad_norm": 0.330078125,
      "learning_rate": 1.164594590280734e-05,
      "loss": 0.845,
      "mean_token_accuracy": 0.7534193771926723,
      "step": 85
    },
    {
      "epoch": 0.5325443786982249,
      "grad_norm": 0.333984375,
      "learning_rate": 1.0619653946285948e-05,
      "loss": 0.8391,
      "mean_token_accuracy": 0.7546228916518883,
      "step": 90
    },
    {
      "epoch": 0.5621301775147929,
      "grad_norm": 0.353515625,
      "learning_rate": 9.586750257511868e-06,
      "loss": 0.8349,
      "mean_token_accuracy": 0.7557750389015123,
      "step": 95
    },
    {
      "epoch": 0.591715976331361,
      "grad_norm": 0.322265625,
      "learning_rate": 8.558255959926533e-06,
      "loss": 0.826,
      "mean_token_accuracy": 0.7576405338899699,
      "step": 100
    },
    {
      "epoch": 0.591715976331361,
      "eval_loss": 0.858128011226654,
      "eval_mean_token_accuracy": 0.7344105051298153,
      "eval_runtime": 2.3145,
      "eval_samples_per_second": 55.735,
      "eval_steps_per_second": 2.16,
      "step": 100
    },
    {
      "epoch": 0.621301775147929,
      "grad_norm": 0.3203125,
      "learning_rate": 7.545145128592009e-06,
      "loss": 0.8279,
      "mean_token_accuracy": 0.7577147828861608,
      "step": 105
    },
    {
      "epoch": 0.650887573964497,
      "grad_norm": 0.314453125,
      "learning_rate": 6.558227696373617e-06,
      "loss": 0.8491,
      "mean_token_accuracy": 0.7520351508799501,
      "step": 110
    },
    {
      "epoch": 0.6804733727810651,
      "grad_norm": 0.322265625,
      "learning_rate": 5.608034111526298e-06,
      "loss": 0.8351,
      "mean_token_accuracy": 0.7557561284875786,
      "step": 115
    },
    {
      "epoch": 0.7100591715976331,
      "grad_norm": 0.328125,
      "learning_rate": 4.704702977392914e-06,
      "loss": 0.8224,
      "mean_token_accuracy": 0.7594720709034851,
      "step": 120
    },
    {
      "epoch": 0.7396449704142012,
      "grad_norm": 0.314453125,
      "learning_rate": 3.857872873103322e-06,
      "loss": 0.8318,
      "mean_token_accuracy": 0.7568701053574788,
      "step": 125
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.326171875,
      "learning_rate": 3.0765795095517026e-06,
      "loss": 0.8362,
      "mean_token_accuracy": 0.755340183100174,
      "step": 130
    },
    {
      "epoch": 0.7988165680473372,
      "grad_norm": 0.3125,
      "learning_rate": 2.369159318001937e-06,
      "loss": 0.8427,
      "mean_token_accuracy": 0.7536929231517968,
      "step": 135
    },
    {
      "epoch": 0.8284023668639053,
      "grad_norm": 0.318359375,
      "learning_rate": 1.743160500034443e-06,
      "loss": 0.834,
      "mean_token_accuracy": 0.756549677941094,
      "step": 140
    },
    {
      "epoch": 0.8579881656804734,
      "grad_norm": 0.328125,
      "learning_rate": 1.2052624879351105e-06,
      "loss": 0.8341,
      "mean_token_accuracy": 0.7563524633707277,
      "step": 145
    },
    {
      "epoch": 0.8875739644970414,
      "grad_norm": 0.30859375,
      "learning_rate": 7.612046748871327e-07,
      "loss": 0.8244,
      "mean_token_accuracy": 0.7588491746092811,
      "step": 150
    },
    {
      "epoch": 0.9171597633136095,
      "grad_norm": 0.31640625,
      "learning_rate": 4.1572517541747294e-07,
      "loss": 0.8148,
      "mean_token_accuracy": 0.7614155323349776,
      "step": 155
    },
    {
      "epoch": 0.9467455621301775,
      "grad_norm": 0.322265625,
      "learning_rate": 1.7251026952640583e-07,
      "loss": 0.8279,
      "mean_token_accuracy": 0.7578688734669882,
      "step": 160
    },
    {
      "epoch": 0.9763313609467456,
      "grad_norm": 0.3125,
      "learning_rate": 3.4155069933301535e-08,
      "loss": 0.8206,
      "mean_token_accuracy": 0.759708155713754,
      "step": 165
    },
    {
      "epoch": 1.0,
      "mean_token_accuracy": 0.754803195434105,
      "step": 169,
      "total_flos": 6.966137809639834e+17,
      "train_loss": 0.8669900795411781,
      "train_runtime": 827.3439,
      "train_samples_per_second": 26.135,
      "train_steps_per_second": 0.204
    }
  ],
  "logging_steps": 5,
  "max_steps": 169,
  "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": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.966137809639834e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}