{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.3902439024390243,
  "eval_steps": 500,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04878048780487805,
      "grad_norm": 5.292644023895264,
      "learning_rate": 1e-05,
      "loss": 4.2989,
      "mean_token_accuracy": 0.7670535296201706,
      "step": 1
    },
    {
      "epoch": 0.0975609756097561,
      "grad_norm": 5.072988033294678,
      "learning_rate": 2e-05,
      "loss": 4.9172,
      "mean_token_accuracy": 0.7377254962921143,
      "step": 2
    },
    {
      "epoch": 0.14634146341463414,
      "grad_norm": 5.262055397033691,
      "learning_rate": 3e-05,
      "loss": 4.3559,
      "mean_token_accuracy": 0.7575394362211227,
      "step": 3
    },
    {
      "epoch": 0.1951219512195122,
      "grad_norm": 4.690796375274658,
      "learning_rate": 4e-05,
      "loss": 4.0699,
      "mean_token_accuracy": 0.775342732667923,
      "step": 4
    },
    {
      "epoch": 0.24390243902439024,
      "grad_norm": 5.084864616394043,
      "learning_rate": 5e-05,
      "loss": 4.0367,
      "mean_token_accuracy": 0.7870949655771255,
      "step": 5
    },
    {
      "epoch": 0.2926829268292683,
      "grad_norm": 5.311507701873779,
      "learning_rate": 4.888888888888889e-05,
      "loss": 4.6084,
      "mean_token_accuracy": 0.7504871934652328,
      "step": 6
    },
    {
      "epoch": 0.34146341463414637,
      "grad_norm": 6.440568447113037,
      "learning_rate": 4.7777777777777784e-05,
      "loss": 4.7291,
      "mean_token_accuracy": 0.7459312677383423,
      "step": 7
    },
    {
      "epoch": 0.3902439024390244,
      "grad_norm": 6.664432048797607,
      "learning_rate": 4.666666666666667e-05,
      "loss": 4.5902,
      "mean_token_accuracy": 0.7366409450769424,
      "step": 8
    },
    {
      "epoch": 0.43902439024390244,
      "grad_norm": 6.393540382385254,
      "learning_rate": 4.555555555555556e-05,
      "loss": 3.9964,
      "mean_token_accuracy": 0.7768445461988449,
      "step": 9
    },
    {
      "epoch": 0.4878048780487805,
      "grad_norm": 7.187014579772949,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 4.8158,
      "mean_token_accuracy": 0.7345478385686874,
      "step": 10
    },
    {
      "epoch": 0.5365853658536586,
      "grad_norm": 6.634271621704102,
      "learning_rate": 4.3333333333333334e-05,
      "loss": 4.3145,
      "mean_token_accuracy": 0.7628218829631805,
      "step": 11
    },
    {
      "epoch": 0.5853658536585366,
      "grad_norm": 6.83073091506958,
      "learning_rate": 4.222222222222222e-05,
      "loss": 4.4571,
      "mean_token_accuracy": 0.7455749660730362,
      "step": 12
    },
    {
      "epoch": 0.6341463414634146,
      "grad_norm": 6.895959854125977,
      "learning_rate": 4.111111111111111e-05,
      "loss": 4.0771,
      "mean_token_accuracy": 0.7768183499574661,
      "step": 13
    },
    {
      "epoch": 0.6829268292682927,
      "grad_norm": 6.867372035980225,
      "learning_rate": 4e-05,
      "loss": 4.4007,
      "mean_token_accuracy": 0.7588072121143341,
      "step": 14
    },
    {
      "epoch": 0.7317073170731707,
      "grad_norm": 6.1517791748046875,
      "learning_rate": 3.888888888888889e-05,
      "loss": 4.1443,
      "mean_token_accuracy": 0.7709105908870697,
      "step": 15
    },
    {
      "epoch": 0.7804878048780488,
      "grad_norm": 6.296253681182861,
      "learning_rate": 3.777777777777778e-05,
      "loss": 4.5376,
      "mean_token_accuracy": 0.7590610980987549,
      "step": 16
    },
    {
      "epoch": 0.8292682926829268,
      "grad_norm": 6.150022983551025,
      "learning_rate": 3.6666666666666666e-05,
      "loss": 4.1576,
      "mean_token_accuracy": 0.7677088677883148,
      "step": 17
    },
    {
      "epoch": 0.8780487804878049,
      "grad_norm": 6.601177215576172,
      "learning_rate": 3.555555555555556e-05,
      "loss": 4.8123,
      "mean_token_accuracy": 0.7380577623844147,
      "step": 18
    },
    {
      "epoch": 0.926829268292683,
      "grad_norm": 6.141186714172363,
      "learning_rate": 3.444444444444445e-05,
      "loss": 4.4355,
      "mean_token_accuracy": 0.7588777989149094,
      "step": 19
    },
    {
      "epoch": 0.975609756097561,
      "grad_norm": 7.759636402130127,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 4.196,
      "mean_token_accuracy": 0.7583473473787308,
      "step": 20
    },
    {
      "epoch": 1.0,
      "grad_norm": 4.481691837310791,
      "learning_rate": 3.222222222222223e-05,
      "loss": 2.7981,
      "mean_token_accuracy": 0.712793231010437,
      "step": 21
    },
    {
      "epoch": 1.048780487804878,
      "grad_norm": 6.39988899230957,
      "learning_rate": 3.111111111111111e-05,
      "loss": 4.1204,
      "mean_token_accuracy": 0.778206542134285,
      "step": 22
    },
    {
      "epoch": 1.0975609756097562,
      "grad_norm": 5.795325756072998,
      "learning_rate": 3e-05,
      "loss": 3.2981,
      "mean_token_accuracy": 0.8128762543201447,
      "step": 23
    },
    {
      "epoch": 1.146341463414634,
      "grad_norm": 5.800813674926758,
      "learning_rate": 2.8888888888888888e-05,
      "loss": 3.6657,
      "mean_token_accuracy": 0.7923942655324936,
      "step": 24
    },
    {
      "epoch": 1.1951219512195121,
      "grad_norm": 6.182375431060791,
      "learning_rate": 2.777777777777778e-05,
      "loss": 4.0254,
      "mean_token_accuracy": 0.773567721247673,
      "step": 25
    },
    {
      "epoch": 1.2439024390243902,
      "grad_norm": 6.226065158843994,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 3.6803,
      "mean_token_accuracy": 0.7928623855113983,
      "step": 26
    },
    {
      "epoch": 1.2926829268292683,
      "grad_norm": 5.555008411407471,
      "learning_rate": 2.5555555555555554e-05,
      "loss": 3.8551,
      "mean_token_accuracy": 0.7828929722309113,
      "step": 27
    },
    {
      "epoch": 1.3414634146341464,
      "grad_norm": 5.869503498077393,
      "learning_rate": 2.4444444444444445e-05,
      "loss": 3.9832,
      "mean_token_accuracy": 0.7784373760223389,
      "step": 28
    },
    {
      "epoch": 1.3902439024390243,
      "grad_norm": 5.494263172149658,
      "learning_rate": 2.3333333333333336e-05,
      "loss": 3.5122,
      "mean_token_accuracy": 0.8014999330043793,
      "step": 29
    },
    {
      "epoch": 1.4390243902439024,
      "grad_norm": 5.7608466148376465,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 3.6381,
      "mean_token_accuracy": 0.7957581877708435,
      "step": 30
    },
    {
      "epoch": 1.4878048780487805,
      "grad_norm": 6.845158576965332,
      "learning_rate": 2.111111111111111e-05,
      "loss": 4.183,
      "mean_token_accuracy": 0.7679397910833359,
      "step": 31
    },
    {
      "epoch": 1.5365853658536586,
      "grad_norm": 6.817462921142578,
      "learning_rate": 2e-05,
      "loss": 4.2954,
      "mean_token_accuracy": 0.7643744349479675,
      "step": 32
    },
    {
      "epoch": 1.5853658536585367,
      "grad_norm": 5.7466349601745605,
      "learning_rate": 1.888888888888889e-05,
      "loss": 3.6042,
      "mean_token_accuracy": 0.7952823042869568,
      "step": 33
    },
    {
      "epoch": 1.6341463414634148,
      "grad_norm": 6.517263412475586,
      "learning_rate": 1.777777777777778e-05,
      "loss": 3.8403,
      "mean_token_accuracy": 0.7832227051258087,
      "step": 34
    },
    {
      "epoch": 1.6829268292682928,
      "grad_norm": 6.389585018157959,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 4.1519,
      "mean_token_accuracy": 0.7758816480636597,
      "step": 35
    },
    {
      "epoch": 1.7317073170731707,
      "grad_norm": 6.375441551208496,
      "learning_rate": 1.5555555555555555e-05,
      "loss": 4.202,
      "mean_token_accuracy": 0.7712285220623016,
      "step": 36
    },
    {
      "epoch": 1.7804878048780488,
      "grad_norm": 6.385025501251221,
      "learning_rate": 1.4444444444444444e-05,
      "loss": 3.7112,
      "mean_token_accuracy": 0.7900592088699341,
      "step": 37
    },
    {
      "epoch": 1.8292682926829267,
      "grad_norm": 6.213353633880615,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 3.9617,
      "mean_token_accuracy": 0.7816430032253265,
      "step": 38
    },
    {
      "epoch": 1.8780487804878048,
      "grad_norm": 6.150025367736816,
      "learning_rate": 1.2222222222222222e-05,
      "loss": 4.0205,
      "mean_token_accuracy": 0.7726276069879532,
      "step": 39
    },
    {
      "epoch": 1.9268292682926829,
      "grad_norm": 6.346643924713135,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 4.3219,
      "mean_token_accuracy": 0.7638045698404312,
      "step": 40
    },
    {
      "epoch": 1.975609756097561,
      "grad_norm": 6.996342182159424,
      "learning_rate": 1e-05,
      "loss": 4.0849,
      "mean_token_accuracy": 0.7685424983501434,
      "step": 41
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.610576629638672,
      "learning_rate": 8.88888888888889e-06,
      "loss": 2.252,
      "mean_token_accuracy": 0.757177472114563,
      "step": 42
    },
    {
      "epoch": 2.048780487804878,
      "grad_norm": 5.402480602264404,
      "learning_rate": 7.777777777777777e-06,
      "loss": 3.6178,
      "mean_token_accuracy": 0.8000039011240005,
      "step": 43
    },
    {
      "epoch": 2.097560975609756,
      "grad_norm": 5.952712535858154,
      "learning_rate": 6.666666666666667e-06,
      "loss": 3.1941,
      "mean_token_accuracy": 0.8220265060663223,
      "step": 44
    },
    {
      "epoch": 2.1463414634146343,
      "grad_norm": 5.641570568084717,
      "learning_rate": 5.555555555555556e-06,
      "loss": 3.7628,
      "mean_token_accuracy": 0.7894951105117798,
      "step": 45
    },
    {
      "epoch": 2.1951219512195124,
      "grad_norm": 5.362001895904541,
      "learning_rate": 4.444444444444445e-06,
      "loss": 3.128,
      "mean_token_accuracy": 0.8270062953233719,
      "step": 46
    },
    {
      "epoch": 2.2439024390243905,
      "grad_norm": 5.856266021728516,
      "learning_rate": 3.3333333333333333e-06,
      "loss": 3.7393,
      "mean_token_accuracy": 0.7911360561847687,
      "step": 47
    },
    {
      "epoch": 2.292682926829268,
      "grad_norm": 6.097686290740967,
      "learning_rate": 2.2222222222222225e-06,
      "loss": 3.452,
      "mean_token_accuracy": 0.8116942197084427,
      "step": 48
    },
    {
      "epoch": 2.341463414634146,
      "grad_norm": 5.644529342651367,
      "learning_rate": 1.1111111111111112e-06,
      "loss": 3.4835,
      "mean_token_accuracy": 0.8041615337133408,
      "step": 49
    },
    {
      "epoch": 2.3902439024390243,
      "grad_norm": 5.310006618499756,
      "learning_rate": 0.0,
      "loss": 3.3661,
      "mean_token_accuracy": 0.8213394284248352,
      "step": 50
    }
  ],
  "logging_steps": 1,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 255785501196288.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}