{
  "best_metric": 0.026573657989501953,
  "best_model_checkpoint": "runs/deepseek_20240423-162824/checkpoint-1000",
  "epoch": 0.31330774653403304,
  "eval_steps": 1000,
  "global_step": 1000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 11.5,
      "learning_rate": 4.0000000000000003e-07,
      "loss": 1.5189,
      "step": 10
    },
    {
      "epoch": 0.01,
      "grad_norm": 16.375,
      "learning_rate": 8.000000000000001e-07,
      "loss": 1.5082,
      "step": 20
    },
    {
      "epoch": 0.01,
      "grad_norm": 21.75,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 1.4715,
      "step": 30
    },
    {
      "epoch": 0.01,
      "grad_norm": 13.8125,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 1.4515,
      "step": 40
    },
    {
      "epoch": 0.02,
      "grad_norm": 9.8125,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.5133,
      "step": 50
    },
    {
      "epoch": 0.02,
      "grad_norm": 10.25,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 1.313,
      "step": 60
    },
    {
      "epoch": 0.02,
      "grad_norm": 18.125,
      "learning_rate": 2.8000000000000003e-06,
      "loss": 1.4489,
      "step": 70
    },
    {
      "epoch": 0.03,
      "grad_norm": 28.0,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 1.2688,
      "step": 80
    },
    {
      "epoch": 0.03,
      "grad_norm": 9.25,
      "learning_rate": 3.6000000000000003e-06,
      "loss": 1.2326,
      "step": 90
    },
    {
      "epoch": 0.03,
      "grad_norm": 25.625,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.1791,
      "step": 100
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.96875,
      "learning_rate": 4.4e-06,
      "loss": 0.9111,
      "step": 110
    },
    {
      "epoch": 0.04,
      "grad_norm": 6.84375,
      "learning_rate": 4.800000000000001e-06,
      "loss": 0.8633,
      "step": 120
    },
    {
      "epoch": 0.04,
      "grad_norm": 7.9375,
      "learning_rate": 5.2e-06,
      "loss": 0.5553,
      "step": 130
    },
    {
      "epoch": 0.04,
      "grad_norm": 14.5625,
      "learning_rate": 5.600000000000001e-06,
      "loss": 0.5916,
      "step": 140
    },
    {
      "epoch": 0.05,
      "grad_norm": 8.8125,
      "learning_rate": 6e-06,
      "loss": 0.4327,
      "step": 150
    },
    {
      "epoch": 0.05,
      "grad_norm": 3.9375,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 0.2799,
      "step": 160
    },
    {
      "epoch": 0.05,
      "grad_norm": 2.34375,
      "learning_rate": 6.800000000000001e-06,
      "loss": 0.2369,
      "step": 170
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.171875,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 0.2997,
      "step": 180
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.287109375,
      "learning_rate": 7.600000000000001e-06,
      "loss": 0.0946,
      "step": 190
    },
    {
      "epoch": 0.06,
      "grad_norm": 18.25,
      "learning_rate": 8.000000000000001e-06,
      "loss": 0.1124,
      "step": 200
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.546875,
      "learning_rate": 8.400000000000001e-06,
      "loss": 0.1131,
      "step": 210
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.71484375,
      "learning_rate": 8.8e-06,
      "loss": 0.0652,
      "step": 220
    },
    {
      "epoch": 0.07,
      "grad_norm": 1.859375,
      "learning_rate": 9.200000000000002e-06,
      "loss": 0.1486,
      "step": 230
    },
    {
      "epoch": 0.08,
      "grad_norm": 1.65625,
      "learning_rate": 9.600000000000001e-06,
      "loss": 0.2169,
      "step": 240
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.1357421875,
      "learning_rate": 1e-05,
      "loss": 0.2408,
      "step": 250
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.055908203125,
      "learning_rate": 1.04e-05,
      "loss": 0.1166,
      "step": 260
    },
    {
      "epoch": 0.08,
      "grad_norm": 5.375,
      "learning_rate": 1.0800000000000002e-05,
      "loss": 0.0114,
      "step": 270
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.0927734375,
      "learning_rate": 1.1200000000000001e-05,
      "loss": 0.1631,
      "step": 280
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.07177734375,
      "learning_rate": 1.16e-05,
      "loss": 0.0603,
      "step": 290
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.482421875,
      "learning_rate": 1.2e-05,
      "loss": 0.1077,
      "step": 300
    },
    {
      "epoch": 0.1,
      "grad_norm": 12.5625,
      "learning_rate": 1.2400000000000002e-05,
      "loss": 0.0832,
      "step": 310
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.048095703125,
      "learning_rate": 1.2800000000000001e-05,
      "loss": 0.0496,
      "step": 320
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.400390625,
      "learning_rate": 1.3200000000000002e-05,
      "loss": 0.1024,
      "step": 330
    },
    {
      "epoch": 0.11,
      "grad_norm": 20.625,
      "learning_rate": 1.3600000000000002e-05,
      "loss": 0.1108,
      "step": 340
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.01611328125,
      "learning_rate": 1.4e-05,
      "loss": 0.0336,
      "step": 350
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.01544189453125,
      "learning_rate": 1.4400000000000001e-05,
      "loss": 0.0165,
      "step": 360
    },
    {
      "epoch": 0.12,
      "grad_norm": 28.0,
      "learning_rate": 1.48e-05,
      "loss": 0.113,
      "step": 370
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.0155029296875,
      "learning_rate": 1.5200000000000002e-05,
      "loss": 0.0221,
      "step": 380
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.0164794921875,
      "learning_rate": 1.5600000000000003e-05,
      "loss": 0.1383,
      "step": 390
    },
    {
      "epoch": 0.13,
      "grad_norm": 1.3203125,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 0.0589,
      "step": 400
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.00909423828125,
      "learning_rate": 1.64e-05,
      "loss": 0.055,
      "step": 410
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.236328125,
      "learning_rate": 1.6800000000000002e-05,
      "loss": 0.0195,
      "step": 420
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.01025390625,
      "learning_rate": 1.72e-05,
      "loss": 0.1477,
      "step": 430
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.007415771484375,
      "learning_rate": 1.76e-05,
      "loss": 0.0708,
      "step": 440
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.01214599609375,
      "learning_rate": 1.8e-05,
      "loss": 0.0808,
      "step": 450
    },
    {
      "epoch": 0.14,
      "grad_norm": 11.3125,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.0689,
      "step": 460
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.013427734375,
      "learning_rate": 1.88e-05,
      "loss": 0.1786,
      "step": 470
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.0113525390625,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.0897,
      "step": 480
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.2490234375,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.0934,
      "step": 490
    },
    {
      "epoch": 0.16,
      "grad_norm": 13.0,
      "learning_rate": 2e-05,
      "loss": 0.1318,
      "step": 500
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.0162353515625,
      "learning_rate": 1.9955555555555557e-05,
      "loss": 0.0156,
      "step": 510
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.0693359375,
      "learning_rate": 1.9911111111111112e-05,
      "loss": 0.067,
      "step": 520
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.007232666015625,
      "learning_rate": 1.9866666666666667e-05,
      "loss": 0.1199,
      "step": 530
    },
    {
      "epoch": 0.17,
      "grad_norm": 1.796875,
      "learning_rate": 1.9822222222222226e-05,
      "loss": 0.0882,
      "step": 540
    },
    {
      "epoch": 0.17,
      "grad_norm": 10.5,
      "learning_rate": 1.977777777777778e-05,
      "loss": 0.1482,
      "step": 550
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.0086669921875,
      "learning_rate": 1.9733333333333336e-05,
      "loss": 0.0937,
      "step": 560
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.328125,
      "learning_rate": 1.968888888888889e-05,
      "loss": 0.0845,
      "step": 570
    },
    {
      "epoch": 0.18,
      "grad_norm": 17.75,
      "learning_rate": 1.9644444444444447e-05,
      "loss": 0.1225,
      "step": 580
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.0068359375,
      "learning_rate": 1.9600000000000002e-05,
      "loss": 0.1117,
      "step": 590
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.00830078125,
      "learning_rate": 1.9555555555555557e-05,
      "loss": 0.1494,
      "step": 600
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.01019287109375,
      "learning_rate": 1.9511111111111113e-05,
      "loss": 0.0575,
      "step": 610
    },
    {
      "epoch": 0.19,
      "grad_norm": 1.5,
      "learning_rate": 1.9466666666666668e-05,
      "loss": 0.0366,
      "step": 620
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.035400390625,
      "learning_rate": 1.9422222222222223e-05,
      "loss": 0.0528,
      "step": 630
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.00860595703125,
      "learning_rate": 1.9377777777777778e-05,
      "loss": 0.0518,
      "step": 640
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.00732421875,
      "learning_rate": 1.9333333333333333e-05,
      "loss": 0.2077,
      "step": 650
    },
    {
      "epoch": 0.21,
      "grad_norm": 3.921875,
      "learning_rate": 1.928888888888889e-05,
      "loss": 0.0559,
      "step": 660
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.01165771484375,
      "learning_rate": 1.9244444444444444e-05,
      "loss": 0.139,
      "step": 670
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.03271484375,
      "learning_rate": 1.9200000000000003e-05,
      "loss": 0.0641,
      "step": 680
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.007232666015625,
      "learning_rate": 1.9155555555555558e-05,
      "loss": 0.0861,
      "step": 690
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.0341796875,
      "learning_rate": 1.9111111111111113e-05,
      "loss": 0.0353,
      "step": 700
    },
    {
      "epoch": 0.22,
      "grad_norm": 6.34375,
      "learning_rate": 1.9066666666666668e-05,
      "loss": 0.0446,
      "step": 710
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.01019287109375,
      "learning_rate": 1.9022222222222223e-05,
      "loss": 0.1009,
      "step": 720
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.007293701171875,
      "learning_rate": 1.897777777777778e-05,
      "loss": 0.0141,
      "step": 730
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.009521484375,
      "learning_rate": 1.8933333333333334e-05,
      "loss": 0.106,
      "step": 740
    },
    {
      "epoch": 0.23,
      "grad_norm": 12.3125,
      "learning_rate": 1.888888888888889e-05,
      "loss": 0.1166,
      "step": 750
    },
    {
      "epoch": 0.24,
      "grad_norm": 12.875,
      "learning_rate": 1.8844444444444444e-05,
      "loss": 0.1105,
      "step": 760
    },
    {
      "epoch": 0.24,
      "grad_norm": 21.25,
      "learning_rate": 1.88e-05,
      "loss": 0.0407,
      "step": 770
    },
    {
      "epoch": 0.24,
      "grad_norm": 18.25,
      "learning_rate": 1.8755555555555558e-05,
      "loss": 0.1127,
      "step": 780
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.032958984375,
      "learning_rate": 1.8711111111111113e-05,
      "loss": 0.022,
      "step": 790
    },
    {
      "epoch": 0.25,
      "grad_norm": 8.6875,
      "learning_rate": 1.866666666666667e-05,
      "loss": 0.1099,
      "step": 800
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.005950927734375,
      "learning_rate": 1.8622222222222224e-05,
      "loss": 0.0456,
      "step": 810
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.005279541015625,
      "learning_rate": 1.857777777777778e-05,
      "loss": 0.086,
      "step": 820
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.03662109375,
      "learning_rate": 1.8533333333333334e-05,
      "loss": 0.0814,
      "step": 830
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.0089111328125,
      "learning_rate": 1.848888888888889e-05,
      "loss": 0.0382,
      "step": 840
    },
    {
      "epoch": 0.27,
      "grad_norm": 292.0,
      "learning_rate": 1.8444444444444448e-05,
      "loss": 0.0244,
      "step": 850
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.01263427734375,
      "learning_rate": 1.8400000000000003e-05,
      "loss": 0.0213,
      "step": 860
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.004608154296875,
      "learning_rate": 1.835555555555556e-05,
      "loss": 0.0438,
      "step": 870
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.005340576171875,
      "learning_rate": 1.8311111111111114e-05,
      "loss": 0.2058,
      "step": 880
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.04931640625,
      "learning_rate": 1.826666666666667e-05,
      "loss": 0.0169,
      "step": 890
    },
    {
      "epoch": 0.28,
      "grad_norm": 9.5,
      "learning_rate": 1.8222222222222224e-05,
      "loss": 0.0693,
      "step": 900
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.00396728515625,
      "learning_rate": 1.817777777777778e-05,
      "loss": 0.0502,
      "step": 910
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.0206298828125,
      "learning_rate": 1.8133333333333335e-05,
      "loss": 0.0479,
      "step": 920
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.046875,
      "learning_rate": 1.808888888888889e-05,
      "loss": 0.0184,
      "step": 930
    },
    {
      "epoch": 0.29,
      "grad_norm": 13.6875,
      "learning_rate": 1.8044444444444445e-05,
      "loss": 0.0315,
      "step": 940
    },
    {
      "epoch": 0.3,
      "grad_norm": 10.0,
      "learning_rate": 1.8e-05,
      "loss": 0.0702,
      "step": 950
    },
    {
      "epoch": 0.3,
      "grad_norm": 7.84375,
      "learning_rate": 1.7955555555555556e-05,
      "loss": 0.162,
      "step": 960
    },
    {
      "epoch": 0.3,
      "grad_norm": 11.25,
      "learning_rate": 1.791111111111111e-05,
      "loss": 0.0772,
      "step": 970
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.0206298828125,
      "learning_rate": 1.7866666666666666e-05,
      "loss": 0.078,
      "step": 980
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.0546875,
      "learning_rate": 1.782222222222222e-05,
      "loss": 0.0804,
      "step": 990
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.053466796875,
      "learning_rate": 1.7777777777777777e-05,
      "loss": 0.0512,
      "step": 1000
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.026573657989501953,
      "eval_runtime": 53.6666,
      "eval_samples_per_second": 18.634,
      "eval_steps_per_second": 18.634,
      "step": 1000
    }
  ],
  "logging_steps": 10,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 1000,
  "total_flos": 1.5733698330624e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}