{
  "best_global_step": 60,
  "best_metric": 0.4450100064277649,
  "best_model_checkpoint": "outputs/checkpoint-60",
  "epoch": 0.08571428571428572,
  "eval_steps": 5,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0014285714285714286,
      "grad_norm": 4.179355621337891,
      "learning_rate": 0.0,
      "loss": 1.3819,
      "step": 1
    },
    {
      "epoch": 0.002857142857142857,
      "grad_norm": 4.235444068908691,
      "learning_rate": 4e-05,
      "loss": 1.5352,
      "step": 2
    },
    {
      "epoch": 0.004285714285714286,
      "grad_norm": 3.5132126808166504,
      "learning_rate": 8e-05,
      "loss": 0.996,
      "step": 3
    },
    {
      "epoch": 0.005714285714285714,
      "grad_norm": 2.1232669353485107,
      "learning_rate": 0.00012,
      "loss": 0.3471,
      "step": 4
    },
    {
      "epoch": 0.007142857142857143,
      "grad_norm": 3.057875633239746,
      "learning_rate": 0.00016,
      "loss": 0.5202,
      "step": 5
    },
    {
      "epoch": 0.007142857142857143,
      "eval_loss": 0.729008674621582,
      "eval_runtime": 61.1607,
      "eval_samples_per_second": 4.578,
      "eval_steps_per_second": 1.145,
      "step": 5
    },
    {
      "epoch": 0.008571428571428572,
      "grad_norm": 5.017824649810791,
      "learning_rate": 0.0002,
      "loss": 0.6374,
      "step": 6
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.8987300395965576,
      "learning_rate": 0.00019555555555555556,
      "loss": 0.7946,
      "step": 7
    },
    {
      "epoch": 0.011428571428571429,
      "grad_norm": 2.3041818141937256,
      "learning_rate": 0.00019111111111111114,
      "loss": 0.475,
      "step": 8
    },
    {
      "epoch": 0.012857142857142857,
      "grad_norm": 1.1344828605651855,
      "learning_rate": 0.0001866666666666667,
      "loss": 0.499,
      "step": 9
    },
    {
      "epoch": 0.014285714285714285,
      "grad_norm": 1.3779399394989014,
      "learning_rate": 0.00018222222222222224,
      "loss": 0.5737,
      "step": 10
    },
    {
      "epoch": 0.014285714285714285,
      "eval_loss": 0.4982975423336029,
      "eval_runtime": 58.9374,
      "eval_samples_per_second": 4.751,
      "eval_steps_per_second": 1.188,
      "step": 10
    },
    {
      "epoch": 0.015714285714285715,
      "grad_norm": 1.039328932762146,
      "learning_rate": 0.00017777777777777779,
      "loss": 0.7145,
      "step": 11
    },
    {
      "epoch": 0.017142857142857144,
      "grad_norm": 1.0184112787246704,
      "learning_rate": 0.00017333333333333334,
      "loss": 0.5086,
      "step": 12
    },
    {
      "epoch": 0.018571428571428572,
      "grad_norm": 3.169090986251831,
      "learning_rate": 0.00016888888888888889,
      "loss": 0.3931,
      "step": 13
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.8889961242675781,
      "learning_rate": 0.00016444444444444444,
      "loss": 0.4175,
      "step": 14
    },
    {
      "epoch": 0.02142857142857143,
      "grad_norm": 1.040206789970398,
      "learning_rate": 0.00016,
      "loss": 0.7937,
      "step": 15
    },
    {
      "epoch": 0.02142857142857143,
      "eval_loss": 0.47482773661613464,
      "eval_runtime": 58.0691,
      "eval_samples_per_second": 4.822,
      "eval_steps_per_second": 1.205,
      "step": 15
    },
    {
      "epoch": 0.022857142857142857,
      "grad_norm": 1.05618417263031,
      "learning_rate": 0.00015555555555555556,
      "loss": 0.4657,
      "step": 16
    },
    {
      "epoch": 0.024285714285714285,
      "grad_norm": 1.636629343032837,
      "learning_rate": 0.0001511111111111111,
      "loss": 0.496,
      "step": 17
    },
    {
      "epoch": 0.025714285714285714,
      "grad_norm": 0.8520965576171875,
      "learning_rate": 0.00014666666666666666,
      "loss": 0.2923,
      "step": 18
    },
    {
      "epoch": 0.027142857142857142,
      "grad_norm": 1.2350469827651978,
      "learning_rate": 0.00014222222222222224,
      "loss": 0.6657,
      "step": 19
    },
    {
      "epoch": 0.02857142857142857,
      "grad_norm": 0.8397138118743896,
      "learning_rate": 0.0001377777777777778,
      "loss": 0.2923,
      "step": 20
    },
    {
      "epoch": 0.02857142857142857,
      "eval_loss": 0.4646180272102356,
      "eval_runtime": 58.7341,
      "eval_samples_per_second": 4.767,
      "eval_steps_per_second": 1.192,
      "step": 20
    },
    {
      "epoch": 0.03,
      "grad_norm": 1.4164972305297852,
      "learning_rate": 0.00013333333333333334,
      "loss": 0.5831,
      "step": 21
    },
    {
      "epoch": 0.03142857142857143,
      "grad_norm": 1.0668251514434814,
      "learning_rate": 0.00012888888888888892,
      "loss": 0.5977,
      "step": 22
    },
    {
      "epoch": 0.032857142857142856,
      "grad_norm": 1.0122352838516235,
      "learning_rate": 0.00012444444444444444,
      "loss": 0.3958,
      "step": 23
    },
    {
      "epoch": 0.03428571428571429,
      "grad_norm": 1.1400679349899292,
      "learning_rate": 0.00012,
      "loss": 0.3899,
      "step": 24
    },
    {
      "epoch": 0.03571428571428571,
      "grad_norm": 1.161012887954712,
      "learning_rate": 0.00011555555555555555,
      "loss": 0.6196,
      "step": 25
    },
    {
      "epoch": 0.03571428571428571,
      "eval_loss": 0.4567541480064392,
      "eval_runtime": 58.108,
      "eval_samples_per_second": 4.819,
      "eval_steps_per_second": 1.205,
      "step": 25
    },
    {
      "epoch": 0.037142857142857144,
      "grad_norm": 1.1181843280792236,
      "learning_rate": 0.00011111111111111112,
      "loss": 0.3504,
      "step": 26
    },
    {
      "epoch": 0.03857142857142857,
      "grad_norm": 1.0887891054153442,
      "learning_rate": 0.00010666666666666667,
      "loss": 0.474,
      "step": 27
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.8779735565185547,
      "learning_rate": 0.00010222222222222222,
      "loss": 0.2359,
      "step": 28
    },
    {
      "epoch": 0.041428571428571426,
      "grad_norm": 1.2299634218215942,
      "learning_rate": 9.777777777777778e-05,
      "loss": 0.5917,
      "step": 29
    },
    {
      "epoch": 0.04285714285714286,
      "grad_norm": 0.6770172715187073,
      "learning_rate": 9.333333333333334e-05,
      "loss": 0.2978,
      "step": 30
    },
    {
      "epoch": 0.04285714285714286,
      "eval_loss": 0.4585675895214081,
      "eval_runtime": 58.9389,
      "eval_samples_per_second": 4.751,
      "eval_steps_per_second": 1.188,
      "step": 30
    },
    {
      "epoch": 0.04428571428571428,
      "grad_norm": 1.2675914764404297,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.464,
      "step": 31
    },
    {
      "epoch": 0.045714285714285714,
      "grad_norm": 0.9487901926040649,
      "learning_rate": 8.444444444444444e-05,
      "loss": 0.6331,
      "step": 32
    },
    {
      "epoch": 0.047142857142857146,
      "grad_norm": 1.002474069595337,
      "learning_rate": 8e-05,
      "loss": 0.5226,
      "step": 33
    },
    {
      "epoch": 0.04857142857142857,
      "grad_norm": 2.1608269214630127,
      "learning_rate": 7.555555555555556e-05,
      "loss": 0.9663,
      "step": 34
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.7519217133522034,
      "learning_rate": 7.111111111111112e-05,
      "loss": 0.2157,
      "step": 35
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.4550328850746155,
      "eval_runtime": 57.9729,
      "eval_samples_per_second": 4.83,
      "eval_steps_per_second": 1.207,
      "step": 35
    },
    {
      "epoch": 0.05142857142857143,
      "grad_norm": 0.7981988787651062,
      "learning_rate": 6.666666666666667e-05,
      "loss": 0.3595,
      "step": 36
    },
    {
      "epoch": 0.05285714285714286,
      "grad_norm": 1.1520148515701294,
      "learning_rate": 6.222222222222222e-05,
      "loss": 0.4654,
      "step": 37
    },
    {
      "epoch": 0.054285714285714284,
      "grad_norm": 0.7494262456893921,
      "learning_rate": 5.7777777777777776e-05,
      "loss": 0.2049,
      "step": 38
    },
    {
      "epoch": 0.055714285714285716,
      "grad_norm": 0.8117587566375732,
      "learning_rate": 5.333333333333333e-05,
      "loss": 0.4791,
      "step": 39
    },
    {
      "epoch": 0.05714285714285714,
      "grad_norm": 0.9275745749473572,
      "learning_rate": 4.888888888888889e-05,
      "loss": 0.4328,
      "step": 40
    },
    {
      "epoch": 0.05714285714285714,
      "eval_loss": 0.44896164536476135,
      "eval_runtime": 59.566,
      "eval_samples_per_second": 4.701,
      "eval_steps_per_second": 1.175,
      "step": 40
    },
    {
      "epoch": 0.05857142857142857,
      "grad_norm": 0.9272159337997437,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.3348,
      "step": 41
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.161618947982788,
      "learning_rate": 4e-05,
      "loss": 0.7938,
      "step": 42
    },
    {
      "epoch": 0.06142857142857143,
      "grad_norm": 0.6889943480491638,
      "learning_rate": 3.555555555555556e-05,
      "loss": 0.3177,
      "step": 43
    },
    {
      "epoch": 0.06285714285714286,
      "grad_norm": 1.577309250831604,
      "learning_rate": 3.111111111111111e-05,
      "loss": 0.5701,
      "step": 44
    },
    {
      "epoch": 0.06428571428571428,
      "grad_norm": 1.2045623064041138,
      "learning_rate": 2.6666666666666667e-05,
      "loss": 0.912,
      "step": 45
    },
    {
      "epoch": 0.06428571428571428,
      "eval_loss": 0.4481067657470703,
      "eval_runtime": 58.0782,
      "eval_samples_per_second": 4.821,
      "eval_steps_per_second": 1.205,
      "step": 45
    },
    {
      "epoch": 0.06571428571428571,
      "grad_norm": 1.6550114154815674,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.8362,
      "step": 46
    },
    {
      "epoch": 0.06714285714285714,
      "grad_norm": 0.6368440985679626,
      "learning_rate": 1.777777777777778e-05,
      "loss": 0.291,
      "step": 47
    },
    {
      "epoch": 0.06857142857142857,
      "grad_norm": 0.8126080632209778,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 0.21,
      "step": 48
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.02597975730896,
      "learning_rate": 8.88888888888889e-06,
      "loss": 0.4724,
      "step": 49
    },
    {
      "epoch": 0.07142857142857142,
      "grad_norm": 0.9014645218849182,
      "learning_rate": 4.444444444444445e-06,
      "loss": 0.351,
      "step": 50
    },
    {
      "epoch": 0.07142857142857142,
      "eval_loss": 0.4469253420829773,
      "eval_runtime": 59.9788,
      "eval_samples_per_second": 4.668,
      "eval_steps_per_second": 1.167,
      "step": 50
    },
    {
      "epoch": 0.07285714285714286,
      "grad_norm": 0.8374767303466797,
      "learning_rate": 0.0,
      "loss": 0.4678,
      "step": 51
    },
    {
      "epoch": 0.07428571428571429,
      "grad_norm": 1.0087125301361084,
      "learning_rate": 3.272727272727273e-05,
      "loss": 0.8477,
      "step": 52
    },
    {
      "epoch": 0.07571428571428572,
      "grad_norm": 0.7490191459655762,
      "learning_rate": 2.909090909090909e-05,
      "loss": 0.2679,
      "step": 53
    },
    {
      "epoch": 0.07714285714285714,
      "grad_norm": 0.8489861488342285,
      "learning_rate": 2.5454545454545454e-05,
      "loss": 0.3833,
      "step": 54
    },
    {
      "epoch": 0.07857142857142857,
      "grad_norm": 0.897487223148346,
      "learning_rate": 2.1818181818181818e-05,
      "loss": 0.5637,
      "step": 55
    },
    {
      "epoch": 0.07857142857142857,
      "eval_loss": 0.44575706124305725,
      "eval_runtime": 57.9488,
      "eval_samples_per_second": 4.832,
      "eval_steps_per_second": 1.208,
      "step": 55
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.9447337985038757,
      "learning_rate": 1.8181818181818182e-05,
      "loss": 0.5945,
      "step": 56
    },
    {
      "epoch": 0.08142857142857143,
      "grad_norm": 0.8487027287483215,
      "learning_rate": 1.4545454545454545e-05,
      "loss": 0.4971,
      "step": 57
    },
    {
      "epoch": 0.08285714285714285,
      "grad_norm": 1.3720009326934814,
      "learning_rate": 1.0909090909090909e-05,
      "loss": 0.905,
      "step": 58
    },
    {
      "epoch": 0.08428571428571428,
      "grad_norm": 0.8870661854743958,
      "learning_rate": 7.272727272727272e-06,
      "loss": 0.4392,
      "step": 59
    },
    {
      "epoch": 0.08571428571428572,
      "grad_norm": 0.8729221224784851,
      "learning_rate": 3.636363636363636e-06,
      "loss": 0.3077,
      "step": 60
    },
    {
      "epoch": 0.08571428571428572,
      "eval_loss": 0.4450100064277649,
      "eval_runtime": 59.8757,
      "eval_samples_per_second": 4.676,
      "eval_steps_per_second": 1.169,
      "step": 60
    }
  ],
  "logging_steps": 1,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3217787051458560.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}