{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0285714285714285,
  "eval_steps": 5,
  "global_step": 53,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.05714285714285714,
      "grad_norm": 12.726211547851562,
      "learning_rate": 1e-05,
      "loss": 16.7409,
      "step": 1
    },
    {
      "epoch": 0.05714285714285714,
      "eval_loss": 9.10863208770752,
      "eval_runtime": 1.8789,
      "eval_samples_per_second": 7.983,
      "eval_steps_per_second": 1.064,
      "step": 1
    },
    {
      "epoch": 0.11428571428571428,
      "grad_norm": 10.197029113769531,
      "learning_rate": 2e-05,
      "loss": 13.2649,
      "step": 2
    },
    {
      "epoch": 0.17142857142857143,
      "grad_norm": 12.611615180969238,
      "learning_rate": 3e-05,
      "loss": 16.1898,
      "step": 3
    },
    {
      "epoch": 0.22857142857142856,
      "grad_norm": 14.422361373901367,
      "learning_rate": 4e-05,
      "loss": 16.7298,
      "step": 4
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 13.265689849853516,
      "learning_rate": 5e-05,
      "loss": 16.0336,
      "step": 5
    },
    {
      "epoch": 0.2857142857142857,
      "eval_loss": 8.701345443725586,
      "eval_runtime": 1.8972,
      "eval_samples_per_second": 7.907,
      "eval_steps_per_second": 1.054,
      "step": 5
    },
    {
      "epoch": 0.34285714285714286,
      "grad_norm": 18.975881576538086,
      "learning_rate": 6e-05,
      "loss": 15.8037,
      "step": 6
    },
    {
      "epoch": 0.4,
      "grad_norm": 19.776723861694336,
      "learning_rate": 7e-05,
      "loss": 14.8698,
      "step": 7
    },
    {
      "epoch": 0.45714285714285713,
      "grad_norm": 13.150531768798828,
      "learning_rate": 8e-05,
      "loss": 12.3575,
      "step": 8
    },
    {
      "epoch": 0.5142857142857142,
      "grad_norm": 15.01668643951416,
      "learning_rate": 9e-05,
      "loss": 10.8048,
      "step": 9
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 14.243483543395996,
      "learning_rate": 0.0001,
      "loss": 9.9134,
      "step": 10
    },
    {
      "epoch": 0.5714285714285714,
      "eval_loss": 4.631707668304443,
      "eval_runtime": 1.8806,
      "eval_samples_per_second": 7.976,
      "eval_steps_per_second": 1.063,
      "step": 10
    },
    {
      "epoch": 0.6285714285714286,
      "grad_norm": 17.032089233398438,
      "learning_rate": 9.986661418317759e-05,
      "loss": 8.326,
      "step": 11
    },
    {
      "epoch": 0.6857142857142857,
      "grad_norm": 31.08342933654785,
      "learning_rate": 9.946716840375551e-05,
      "loss": 6.0208,
      "step": 12
    },
    {
      "epoch": 0.7428571428571429,
      "grad_norm": 19.09461212158203,
      "learning_rate": 9.880379387779637e-05,
      "loss": 5.0303,
      "step": 13
    },
    {
      "epoch": 0.8,
      "grad_norm": 20.249990463256836,
      "learning_rate": 9.78800299954203e-05,
      "loss": 3.0912,
      "step": 14
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 12.805388450622559,
      "learning_rate": 9.67008054366274e-05,
      "loss": 1.385,
      "step": 15
    },
    {
      "epoch": 0.8571428571428571,
      "eval_loss": 1.539831519126892,
      "eval_runtime": 1.8804,
      "eval_samples_per_second": 7.977,
      "eval_steps_per_second": 1.064,
      "step": 15
    },
    {
      "epoch": 0.9142857142857143,
      "grad_norm": 16.641372680664062,
      "learning_rate": 9.527241187465734e-05,
      "loss": 1.2639,
      "step": 16
    },
    {
      "epoch": 0.9714285714285714,
      "grad_norm": 16.987781524658203,
      "learning_rate": 9.360247040719039e-05,
      "loss": 1.4489,
      "step": 17
    },
    {
      "epoch": 1.0285714285714285,
      "grad_norm": 10.107866287231445,
      "learning_rate": 9.16998908944939e-05,
      "loss": 1.8918,
      "step": 18
    },
    {
      "epoch": 1.0857142857142856,
      "grad_norm": 14.143921852111816,
      "learning_rate": 8.957482442146272e-05,
      "loss": 0.9066,
      "step": 19
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 6.628956317901611,
      "learning_rate": 8.72386091371891e-05,
      "loss": 1.3674,
      "step": 20
    },
    {
      "epoch": 1.1428571428571428,
      "eval_loss": 1.024061918258667,
      "eval_runtime": 1.8791,
      "eval_samples_per_second": 7.983,
      "eval_steps_per_second": 1.064,
      "step": 20
    },
    {
      "epoch": 1.2,
      "grad_norm": 13.004719734191895,
      "learning_rate": 8.47037097610317e-05,
      "loss": 0.966,
      "step": 21
    },
    {
      "epoch": 1.2571428571428571,
      "grad_norm": 9.043590545654297,
      "learning_rate": 8.198365107794457e-05,
      "loss": 1.1478,
      "step": 22
    },
    {
      "epoch": 1.3142857142857143,
      "grad_norm": 8.619237899780273,
      "learning_rate": 7.909294577789766e-05,
      "loss": 0.3326,
      "step": 23
    },
    {
      "epoch": 1.3714285714285714,
      "grad_norm": 7.541545867919922,
      "learning_rate": 7.604701702439651e-05,
      "loss": 0.251,
      "step": 24
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 10.902060508728027,
      "learning_rate": 7.286211616523193e-05,
      "loss": 0.8831,
      "step": 25
    },
    {
      "epoch": 1.4285714285714286,
      "eval_loss": 0.7456358671188354,
      "eval_runtime": 1.8792,
      "eval_samples_per_second": 7.982,
      "eval_steps_per_second": 1.064,
      "step": 25
    },
    {
      "epoch": 1.4857142857142858,
      "grad_norm": 9.19676399230957,
      "learning_rate": 6.95552360245078e-05,
      "loss": 0.728,
      "step": 26
    },
    {
      "epoch": 1.5428571428571427,
      "grad_norm": 12.570167541503906,
      "learning_rate": 6.614402023857232e-05,
      "loss": 0.92,
      "step": 27
    },
    {
      "epoch": 1.6,
      "grad_norm": 8.061638832092285,
      "learning_rate": 6.264666911958404e-05,
      "loss": 0.3867,
      "step": 28
    },
    {
      "epoch": 1.657142857142857,
      "grad_norm": 5.505425453186035,
      "learning_rate": 5.908184254897182e-05,
      "loss": 0.4116,
      "step": 29
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 4.206918239593506,
      "learning_rate": 5.546856041889373e-05,
      "loss": 0.26,
      "step": 30
    },
    {
      "epoch": 1.7142857142857144,
      "eval_loss": 0.6001853942871094,
      "eval_runtime": 1.8809,
      "eval_samples_per_second": 7.975,
      "eval_steps_per_second": 1.063,
      "step": 30
    },
    {
      "epoch": 1.7714285714285714,
      "grad_norm": 7.199727535247803,
      "learning_rate": 5.182610115288295e-05,
      "loss": 0.6765,
      "step": 31
    },
    {
      "epoch": 1.8285714285714287,
      "grad_norm": 5.864309310913086,
      "learning_rate": 4.817389884711705e-05,
      "loss": 0.6613,
      "step": 32
    },
    {
      "epoch": 1.8857142857142857,
      "grad_norm": 2.527921676635742,
      "learning_rate": 4.4531439581106295e-05,
      "loss": 0.0827,
      "step": 33
    },
    {
      "epoch": 1.9428571428571428,
      "grad_norm": 5.816792011260986,
      "learning_rate": 4.0918157451028185e-05,
      "loss": 0.2566,
      "step": 34
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.379115104675293,
      "learning_rate": 3.735333088041596e-05,
      "loss": 0.3601,
      "step": 35
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.6280831694602966,
      "eval_runtime": 1.8813,
      "eval_samples_per_second": 7.973,
      "eval_steps_per_second": 1.063,
      "step": 35
    },
    {
      "epoch": 2.057142857142857,
      "grad_norm": 5.1028642654418945,
      "learning_rate": 3.38559797614277e-05,
      "loss": 0.1492,
      "step": 36
    },
    {
      "epoch": 2.1142857142857143,
      "grad_norm": 1.6019138097763062,
      "learning_rate": 3.0444763975492208e-05,
      "loss": 0.058,
      "step": 37
    },
    {
      "epoch": 2.1714285714285713,
      "grad_norm": 3.3319895267486572,
      "learning_rate": 2.7137883834768073e-05,
      "loss": 0.1912,
      "step": 38
    },
    {
      "epoch": 2.2285714285714286,
      "grad_norm": 3.3347697257995605,
      "learning_rate": 2.3952982975603496e-05,
      "loss": 0.1467,
      "step": 39
    },
    {
      "epoch": 2.2857142857142856,
      "grad_norm": 4.849230766296387,
      "learning_rate": 2.090705422210237e-05,
      "loss": 0.2343,
      "step": 40
    },
    {
      "epoch": 2.2857142857142856,
      "eval_loss": 0.6117228865623474,
      "eval_runtime": 1.8795,
      "eval_samples_per_second": 7.981,
      "eval_steps_per_second": 1.064,
      "step": 40
    },
    {
      "epoch": 2.342857142857143,
      "grad_norm": 0.7933540344238281,
      "learning_rate": 1.801634892205545e-05,
      "loss": 0.0241,
      "step": 41
    },
    {
      "epoch": 2.4,
      "grad_norm": 4.947017192840576,
      "learning_rate": 1.5296290238968303e-05,
      "loss": 0.1974,
      "step": 42
    },
    {
      "epoch": 2.4571428571428573,
      "grad_norm": 4.42537784576416,
      "learning_rate": 1.2761390862810907e-05,
      "loss": 0.2682,
      "step": 43
    },
    {
      "epoch": 2.5142857142857142,
      "grad_norm": 2.0344936847686768,
      "learning_rate": 1.0425175578537299e-05,
      "loss": 0.1215,
      "step": 44
    },
    {
      "epoch": 2.571428571428571,
      "grad_norm": 2.9451773166656494,
      "learning_rate": 8.30010910550611e-06,
      "loss": 0.1397,
      "step": 45
    },
    {
      "epoch": 2.571428571428571,
      "eval_loss": 0.567788302898407,
      "eval_runtime": 1.8814,
      "eval_samples_per_second": 7.973,
      "eval_steps_per_second": 1.063,
      "step": 45
    },
    {
      "epoch": 2.6285714285714286,
      "grad_norm": 4.362372398376465,
      "learning_rate": 6.397529592809614e-06,
      "loss": 0.1807,
      "step": 46
    },
    {
      "epoch": 2.685714285714286,
      "grad_norm": 1.7142335176467896,
      "learning_rate": 4.727588125342669e-06,
      "loss": 0.0381,
      "step": 47
    },
    {
      "epoch": 2.742857142857143,
      "grad_norm": 5.089944362640381,
      "learning_rate": 3.299194563372604e-06,
      "loss": 0.7555,
      "step": 48
    },
    {
      "epoch": 2.8,
      "grad_norm": 2.963768482208252,
      "learning_rate": 2.1199700045797077e-06,
      "loss": 0.0599,
      "step": 49
    },
    {
      "epoch": 2.857142857142857,
      "grad_norm": 3.1674277782440186,
      "learning_rate": 1.196206122203647e-06,
      "loss": 0.0929,
      "step": 50
    },
    {
      "epoch": 2.857142857142857,
      "eval_loss": 0.5395380854606628,
      "eval_runtime": 1.879,
      "eval_samples_per_second": 7.983,
      "eval_steps_per_second": 1.064,
      "step": 50
    },
    {
      "epoch": 2.914285714285714,
      "grad_norm": 3.3688862323760986,
      "learning_rate": 5.328315962444874e-07,
      "loss": 0.3766,
      "step": 51
    },
    {
      "epoch": 2.9714285714285715,
      "grad_norm": 2.4611260890960693,
      "learning_rate": 1.333858168224178e-07,
      "loss": 0.0894,
      "step": 52
    },
    {
      "epoch": 3.0285714285714285,
      "grad_norm": 2.989964485168457,
      "learning_rate": 0.0,
      "loss": 0.0514,
      "step": 53
    }
  ],
  "logging_steps": 1,
  "max_steps": 53,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.4633626150764544e+16,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}