{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 27080,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.36927621861152143,
      "grad_norm": 13247.388671875,
      "learning_rate": 0.0005993999999999999,
      "loss": 0.6625,
      "step": 1000
    },
    {
      "epoch": 0.7385524372230429,
      "grad_norm": 9940.6201171875,
      "learning_rate": 0.0005770168711656441,
      "loss": 0.4984,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.46332530241587166,
      "eval_loss": 1.8212859630584717,
      "eval_runtime": 10.2423,
      "eval_samples_per_second": 60.826,
      "eval_steps_per_second": 1.953,
      "step": 2708
    },
    {
      "epoch": 1.1078286558345642,
      "grad_norm": 8275.166015625,
      "learning_rate": 0.000554010736196319,
      "loss": 0.4532,
      "step": 3000
    },
    {
      "epoch": 1.4771048744460857,
      "grad_norm": 7695.97265625,
      "learning_rate": 0.0005310046012269938,
      "loss": 0.4253,
      "step": 4000
    },
    {
      "epoch": 1.846381093057607,
      "grad_norm": 6635.02685546875,
      "learning_rate": 0.0005079984662576687,
      "loss": 0.408,
      "step": 5000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5106941037150584,
      "eval_loss": 1.6397608518600464,
      "eval_runtime": 9.7196,
      "eval_samples_per_second": 64.097,
      "eval_steps_per_second": 2.058,
      "step": 5416
    },
    {
      "epoch": 2.2156573116691285,
      "grad_norm": 7749.61181640625,
      "learning_rate": 0.00048499233128834355,
      "loss": 0.3934,
      "step": 6000
    },
    {
      "epoch": 2.58493353028065,
      "grad_norm": 6143.23388671875,
      "learning_rate": 0.0004619861963190184,
      "loss": 0.3836,
      "step": 7000
    },
    {
      "epoch": 2.9542097488921715,
      "grad_norm": 6868.88916015625,
      "learning_rate": 0.00043898006134969323,
      "loss": 0.3764,
      "step": 8000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5333796131966716,
      "eval_loss": 1.5575107336044312,
      "eval_runtime": 9.7301,
      "eval_samples_per_second": 64.028,
      "eval_steps_per_second": 2.055,
      "step": 8124
    },
    {
      "epoch": 3.323485967503693,
      "grad_norm": 6560.751953125,
      "learning_rate": 0.00041597392638036807,
      "loss": 0.3684,
      "step": 9000
    },
    {
      "epoch": 3.692762186115214,
      "grad_norm": 6382.58203125,
      "learning_rate": 0.0003929677914110429,
      "loss": 0.3628,
      "step": 10000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.547043690494514,
      "eval_loss": 1.5047563314437866,
      "eval_runtime": 9.717,
      "eval_samples_per_second": 64.115,
      "eval_steps_per_second": 2.058,
      "step": 10832
    },
    {
      "epoch": 4.062038404726736,
      "grad_norm": 5827.2607421875,
      "learning_rate": 0.0003699616564417178,
      "loss": 0.3585,
      "step": 11000
    },
    {
      "epoch": 4.431314623338257,
      "grad_norm": 6025.1650390625,
      "learning_rate": 0.00034695552147239264,
      "loss": 0.3539,
      "step": 12000
    },
    {
      "epoch": 4.800590841949779,
      "grad_norm": 5609.11474609375,
      "learning_rate": 0.0003239493865030675,
      "loss": 0.3505,
      "step": 13000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5583267630586173,
      "eval_loss": 1.4623572826385498,
      "eval_runtime": 9.7131,
      "eval_samples_per_second": 64.14,
      "eval_steps_per_second": 2.059,
      "step": 13540
    },
    {
      "epoch": 5.1698670605613,
      "grad_norm": 5670.53466796875,
      "learning_rate": 0.0003009432515337423,
      "loss": 0.3465,
      "step": 14000
    },
    {
      "epoch": 5.539143279172821,
      "grad_norm": 6664.6083984375,
      "learning_rate": 0.00027793711656441715,
      "loss": 0.3426,
      "step": 15000
    },
    {
      "epoch": 5.908419497784343,
      "grad_norm": 5798.9404296875,
      "learning_rate": 0.000254930981595092,
      "loss": 0.3399,
      "step": 16000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5711521487154196,
      "eval_loss": 1.427840232849121,
      "eval_runtime": 9.718,
      "eval_samples_per_second": 64.108,
      "eval_steps_per_second": 2.058,
      "step": 16248
    },
    {
      "epoch": 6.277695716395864,
      "grad_norm": 5907.61376953125,
      "learning_rate": 0.00023192484662576683,
      "loss": 0.3356,
      "step": 17000
    },
    {
      "epoch": 6.646971935007386,
      "grad_norm": 5923.216796875,
      "learning_rate": 0.0002089187116564417,
      "loss": 0.3334,
      "step": 18000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5765989326313872,
      "eval_loss": 1.406113624572754,
      "eval_runtime": 9.7146,
      "eval_samples_per_second": 64.13,
      "eval_steps_per_second": 2.059,
      "step": 18956
    },
    {
      "epoch": 7.016248153618907,
      "grad_norm": 6729.748046875,
      "learning_rate": 0.00018591257668711654,
      "loss": 0.3315,
      "step": 19000
    },
    {
      "epoch": 7.385524372230428,
      "grad_norm": 6439.97607421875,
      "learning_rate": 0.00016290644171779137,
      "loss": 0.3267,
      "step": 20000
    },
    {
      "epoch": 7.75480059084195,
      "grad_norm": 6263.06689453125,
      "learning_rate": 0.00013990030674846624,
      "loss": 0.3248,
      "step": 21000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5833838537723848,
      "eval_loss": 1.3805782794952393,
      "eval_runtime": 9.8981,
      "eval_samples_per_second": 62.941,
      "eval_steps_per_second": 2.021,
      "step": 21664
    },
    {
      "epoch": 8.124076809453472,
      "grad_norm": 6769.88916015625,
      "learning_rate": 0.00011689417177914109,
      "loss": 0.3226,
      "step": 22000
    },
    {
      "epoch": 8.493353028064993,
      "grad_norm": 5885.490234375,
      "learning_rate": 9.388803680981594e-05,
      "loss": 0.3188,
      "step": 23000
    },
    {
      "epoch": 8.862629246676514,
      "grad_norm": 6447.85498046875,
      "learning_rate": 7.08819018404908e-05,
      "loss": 0.3178,
      "step": 24000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5892798245972238,
      "eval_loss": 1.3639582395553589,
      "eval_runtime": 9.9358,
      "eval_samples_per_second": 62.703,
      "eval_steps_per_second": 2.013,
      "step": 24372
    },
    {
      "epoch": 9.231905465288035,
      "grad_norm": 6391.68896484375,
      "learning_rate": 4.787576687116564e-05,
      "loss": 0.3141,
      "step": 25000
    },
    {
      "epoch": 9.601181683899556,
      "grad_norm": 6642.7880859375,
      "learning_rate": 2.4869631901840487e-05,
      "loss": 0.3118,
      "step": 26000
    },
    {
      "epoch": 9.970457902511079,
      "grad_norm": 6367.48193359375,
      "learning_rate": 1.863496932515337e-06,
      "loss": 0.311,
      "step": 27000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5924084271233505,
      "eval_loss": 1.3527692556381226,
      "eval_runtime": 9.7249,
      "eval_samples_per_second": 64.063,
      "eval_steps_per_second": 2.057,
      "step": 27080
    },
    {
      "epoch": 10.0,
      "step": 27080,
      "total_flos": 2.2639125528576e+17,
      "train_loss": 0.36916396628454484,
      "train_runtime": 23461.5127,
      "train_samples_per_second": 36.93,
      "train_steps_per_second": 1.154
    }
  ],
  "logging_steps": 1000,
  "max_steps": 27080,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 2.2639125528576e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}