{
  "best_metric": 0.39595454931259155,
  "best_model_checkpoint": "mikhail-panzo/ceb_b128_le5_s4000/checkpoint-3000",
  "epoch": 235.2941176470588,
  "eval_steps": 500,
  "global_step": 3000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.9215686274509802,
      "grad_norm": 4.645490646362305,
      "learning_rate": 2.4500000000000004e-07,
      "loss": 0.7929,
      "step": 50
    },
    {
      "epoch": 7.8431372549019605,
      "grad_norm": 1.4075939655303955,
      "learning_rate": 4.95e-07,
      "loss": 0.7648,
      "step": 100
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 4.839516639709473,
      "learning_rate": 7.450000000000001e-07,
      "loss": 0.7575,
      "step": 150
    },
    {
      "epoch": 15.686274509803921,
      "grad_norm": 1.0808578729629517,
      "learning_rate": 9.950000000000002e-07,
      "loss": 0.7118,
      "step": 200
    },
    {
      "epoch": 19.607843137254903,
      "grad_norm": 2.4112303256988525,
      "learning_rate": 1.2450000000000002e-06,
      "loss": 0.706,
      "step": 250
    },
    {
      "epoch": 23.529411764705884,
      "grad_norm": 0.9267112016677856,
      "learning_rate": 1.495e-06,
      "loss": 0.6756,
      "step": 300
    },
    {
      "epoch": 27.45098039215686,
      "grad_norm": 1.3859704732894897,
      "learning_rate": 1.745e-06,
      "loss": 0.6412,
      "step": 350
    },
    {
      "epoch": 31.372549019607842,
      "grad_norm": 0.9298150539398193,
      "learning_rate": 1.9950000000000004e-06,
      "loss": 0.5781,
      "step": 400
    },
    {
      "epoch": 35.294117647058826,
      "grad_norm": 1.0941523313522339,
      "learning_rate": 2.245e-06,
      "loss": 0.5381,
      "step": 450
    },
    {
      "epoch": 39.21568627450981,
      "grad_norm": 1.264251708984375,
      "learning_rate": 2.4950000000000003e-06,
      "loss": 0.5272,
      "step": 500
    },
    {
      "epoch": 39.21568627450981,
      "eval_loss": 0.4584241807460785,
      "eval_runtime": 6.3895,
      "eval_samples_per_second": 28.171,
      "eval_steps_per_second": 3.6,
      "step": 500
    },
    {
      "epoch": 43.13725490196079,
      "grad_norm": 0.8968011736869812,
      "learning_rate": 2.7450000000000004e-06,
      "loss": 0.5121,
      "step": 550
    },
    {
      "epoch": 47.05882352941177,
      "grad_norm": 1.1953343152999878,
      "learning_rate": 2.995e-06,
      "loss": 0.5047,
      "step": 600
    },
    {
      "epoch": 50.98039215686274,
      "grad_norm": 0.8916338682174683,
      "learning_rate": 3.2450000000000003e-06,
      "loss": 0.5013,
      "step": 650
    },
    {
      "epoch": 54.90196078431372,
      "grad_norm": 1.166737675666809,
      "learning_rate": 3.495e-06,
      "loss": 0.4847,
      "step": 700
    },
    {
      "epoch": 58.8235294117647,
      "grad_norm": 0.9191999435424805,
      "learning_rate": 3.745e-06,
      "loss": 0.4849,
      "step": 750
    },
    {
      "epoch": 62.745098039215684,
      "grad_norm": 1.0393210649490356,
      "learning_rate": 3.995000000000001e-06,
      "loss": 0.4827,
      "step": 800
    },
    {
      "epoch": 66.66666666666667,
      "grad_norm": 0.764959454536438,
      "learning_rate": 4.245e-06,
      "loss": 0.483,
      "step": 850
    },
    {
      "epoch": 70.58823529411765,
      "grad_norm": 1.0713616609573364,
      "learning_rate": 4.495e-06,
      "loss": 0.4762,
      "step": 900
    },
    {
      "epoch": 74.50980392156863,
      "grad_norm": 0.8433477282524109,
      "learning_rate": 4.745e-06,
      "loss": 0.4702,
      "step": 950
    },
    {
      "epoch": 78.43137254901961,
      "grad_norm": 0.6966288685798645,
      "learning_rate": 4.9950000000000005e-06,
      "loss": 0.4634,
      "step": 1000
    },
    {
      "epoch": 78.43137254901961,
      "eval_loss": 0.42202475666999817,
      "eval_runtime": 6.3604,
      "eval_samples_per_second": 28.3,
      "eval_steps_per_second": 3.616,
      "step": 1000
    },
    {
      "epoch": 82.3529411764706,
      "grad_norm": 0.9115990400314331,
      "learning_rate": 5.245e-06,
      "loss": 0.47,
      "step": 1050
    },
    {
      "epoch": 86.27450980392157,
      "grad_norm": 0.6754831075668335,
      "learning_rate": 5.495000000000001e-06,
      "loss": 0.4605,
      "step": 1100
    },
    {
      "epoch": 90.19607843137256,
      "grad_norm": 1.0708327293395996,
      "learning_rate": 5.745000000000001e-06,
      "loss": 0.458,
      "step": 1150
    },
    {
      "epoch": 94.11764705882354,
      "grad_norm": 0.7757265567779541,
      "learning_rate": 5.995000000000001e-06,
      "loss": 0.456,
      "step": 1200
    },
    {
      "epoch": 98.03921568627452,
      "grad_norm": 1.1435647010803223,
      "learning_rate": 6.245000000000001e-06,
      "loss": 0.4576,
      "step": 1250
    },
    {
      "epoch": 101.96078431372548,
      "grad_norm": 0.8143028020858765,
      "learning_rate": 6.4950000000000005e-06,
      "loss": 0.4518,
      "step": 1300
    },
    {
      "epoch": 105.88235294117646,
      "grad_norm": 0.8940721750259399,
      "learning_rate": 6.745000000000001e-06,
      "loss": 0.4515,
      "step": 1350
    },
    {
      "epoch": 109.80392156862744,
      "grad_norm": 1.8656580448150635,
      "learning_rate": 6.995000000000001e-06,
      "loss": 0.4516,
      "step": 1400
    },
    {
      "epoch": 113.72549019607843,
      "grad_norm": 0.7817286252975464,
      "learning_rate": 7.245000000000001e-06,
      "loss": 0.4412,
      "step": 1450
    },
    {
      "epoch": 117.6470588235294,
      "grad_norm": 1.806294322013855,
      "learning_rate": 7.495000000000001e-06,
      "loss": 0.4466,
      "step": 1500
    },
    {
      "epoch": 117.6470588235294,
      "eval_loss": 0.41040292382240295,
      "eval_runtime": 6.4965,
      "eval_samples_per_second": 27.707,
      "eval_steps_per_second": 3.54,
      "step": 1500
    },
    {
      "epoch": 121.56862745098039,
      "grad_norm": 0.6831104755401611,
      "learning_rate": 7.745e-06,
      "loss": 0.4461,
      "step": 1550
    },
    {
      "epoch": 125.49019607843137,
      "grad_norm": 1.195868968963623,
      "learning_rate": 7.995e-06,
      "loss": 0.4429,
      "step": 1600
    },
    {
      "epoch": 129.41176470588235,
      "grad_norm": 1.1746853590011597,
      "learning_rate": 8.245000000000002e-06,
      "loss": 0.4358,
      "step": 1650
    },
    {
      "epoch": 133.33333333333334,
      "grad_norm": 1.2797439098358154,
      "learning_rate": 8.495e-06,
      "loss": 0.4383,
      "step": 1700
    },
    {
      "epoch": 137.2549019607843,
      "grad_norm": 0.6744837760925293,
      "learning_rate": 8.745000000000002e-06,
      "loss": 0.4416,
      "step": 1750
    },
    {
      "epoch": 141.1764705882353,
      "grad_norm": 0.7655614018440247,
      "learning_rate": 8.995000000000001e-06,
      "loss": 0.4338,
      "step": 1800
    },
    {
      "epoch": 145.09803921568627,
      "grad_norm": 0.9920282363891602,
      "learning_rate": 9.245e-06,
      "loss": 0.4337,
      "step": 1850
    },
    {
      "epoch": 149.01960784313727,
      "grad_norm": 0.9740642309188843,
      "learning_rate": 9.495000000000001e-06,
      "loss": 0.4309,
      "step": 1900
    },
    {
      "epoch": 152.94117647058823,
      "grad_norm": 0.9331285953521729,
      "learning_rate": 9.745e-06,
      "loss": 0.4337,
      "step": 1950
    },
    {
      "epoch": 156.86274509803923,
      "grad_norm": 0.8512988686561584,
      "learning_rate": 9.995000000000002e-06,
      "loss": 0.4289,
      "step": 2000
    },
    {
      "epoch": 156.86274509803923,
      "eval_loss": 0.4016592502593994,
      "eval_runtime": 6.4392,
      "eval_samples_per_second": 27.954,
      "eval_steps_per_second": 3.572,
      "step": 2000
    },
    {
      "epoch": 160.7843137254902,
      "grad_norm": 0.7746613025665283,
      "learning_rate": 9.755e-06,
      "loss": 0.4306,
      "step": 2050
    },
    {
      "epoch": 164.7058823529412,
      "grad_norm": 0.6868831515312195,
      "learning_rate": 9.505000000000001e-06,
      "loss": 0.4302,
      "step": 2100
    },
    {
      "epoch": 168.62745098039215,
      "grad_norm": 1.010834813117981,
      "learning_rate": 9.255e-06,
      "loss": 0.4254,
      "step": 2150
    },
    {
      "epoch": 172.54901960784315,
      "grad_norm": 1.054592490196228,
      "learning_rate": 9.005000000000001e-06,
      "loss": 0.4248,
      "step": 2200
    },
    {
      "epoch": 176.47058823529412,
      "grad_norm": 0.8121660351753235,
      "learning_rate": 8.755e-06,
      "loss": 0.4227,
      "step": 2250
    },
    {
      "epoch": 180.3921568627451,
      "grad_norm": 0.6637047529220581,
      "learning_rate": 8.505e-06,
      "loss": 0.4232,
      "step": 2300
    },
    {
      "epoch": 184.31372549019608,
      "grad_norm": 1.0822277069091797,
      "learning_rate": 8.255000000000001e-06,
      "loss": 0.4226,
      "step": 2350
    },
    {
      "epoch": 188.23529411764707,
      "grad_norm": 0.759693443775177,
      "learning_rate": 8.005e-06,
      "loss": 0.4236,
      "step": 2400
    },
    {
      "epoch": 192.15686274509804,
      "grad_norm": 0.576042652130127,
      "learning_rate": 7.755000000000001e-06,
      "loss": 0.4162,
      "step": 2450
    },
    {
      "epoch": 196.07843137254903,
      "grad_norm": 0.8360034227371216,
      "learning_rate": 7.505e-06,
      "loss": 0.4223,
      "step": 2500
    },
    {
      "epoch": 196.07843137254903,
      "eval_loss": 0.39692553877830505,
      "eval_runtime": 6.4387,
      "eval_samples_per_second": 27.956,
      "eval_steps_per_second": 3.572,
      "step": 2500
    },
    {
      "epoch": 200.0,
      "grad_norm": 0.7426376342773438,
      "learning_rate": 7.255000000000001e-06,
      "loss": 0.4157,
      "step": 2550
    },
    {
      "epoch": 203.92156862745097,
      "grad_norm": 1.1800576448440552,
      "learning_rate": 7.005000000000001e-06,
      "loss": 0.419,
      "step": 2600
    },
    {
      "epoch": 207.84313725490196,
      "grad_norm": 0.7355245351791382,
      "learning_rate": 6.7550000000000005e-06,
      "loss": 0.4174,
      "step": 2650
    },
    {
      "epoch": 211.76470588235293,
      "grad_norm": 0.5805600881576538,
      "learning_rate": 6.505e-06,
      "loss": 0.4146,
      "step": 2700
    },
    {
      "epoch": 215.68627450980392,
      "grad_norm": 0.9223101139068604,
      "learning_rate": 6.255e-06,
      "loss": 0.4178,
      "step": 2750
    },
    {
      "epoch": 219.6078431372549,
      "grad_norm": 0.8155106902122498,
      "learning_rate": 6.005000000000001e-06,
      "loss": 0.4151,
      "step": 2800
    },
    {
      "epoch": 223.52941176470588,
      "grad_norm": 0.6420881748199463,
      "learning_rate": 5.755000000000001e-06,
      "loss": 0.4156,
      "step": 2850
    },
    {
      "epoch": 227.45098039215685,
      "grad_norm": 0.7704824209213257,
      "learning_rate": 5.505000000000001e-06,
      "loss": 0.4157,
      "step": 2900
    },
    {
      "epoch": 231.37254901960785,
      "grad_norm": 0.6147534251213074,
      "learning_rate": 5.2550000000000005e-06,
      "loss": 0.4177,
      "step": 2950
    },
    {
      "epoch": 235.2941176470588,
      "grad_norm": 0.758510172367096,
      "learning_rate": 5.0049999999999995e-06,
      "loss": 0.4149,
      "step": 3000
    },
    {
      "epoch": 235.2941176470588,
      "eval_loss": 0.39595454931259155,
      "eval_runtime": 6.3959,
      "eval_samples_per_second": 28.143,
      "eval_steps_per_second": 3.596,
      "step": 3000
    }
  ],
  "logging_steps": 50,
  "max_steps": 4000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 334,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5283263362154376e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}