{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.974948758824869,
  "global_step": 10950,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23,
      "learning_rate": 4.553734061930783e-05,
      "loss": 1.3816,
      "step": 250
    },
    {
      "epoch": 0.23,
      "eval_loss": 0.1994573473930359,
      "eval_runtime": 49.3939,
      "eval_samples_per_second": 50.553,
      "eval_steps_per_second": 6.337,
      "step": 250
    },
    {
      "epoch": 0.46,
      "learning_rate": 9.107468123861566e-05,
      "loss": 0.1818,
      "step": 500
    },
    {
      "epoch": 0.46,
      "eval_loss": 0.18140748143196106,
      "eval_runtime": 49.3506,
      "eval_samples_per_second": 50.597,
      "eval_steps_per_second": 6.342,
      "step": 500
    },
    {
      "epoch": 0.68,
      "learning_rate": 9.807120237981e-05,
      "loss": 0.1716,
      "step": 750
    },
    {
      "epoch": 0.68,
      "eval_loss": 0.1712454855442047,
      "eval_runtime": 49.3274,
      "eval_samples_per_second": 50.621,
      "eval_steps_per_second": 6.345,
      "step": 750
    },
    {
      "epoch": 0.91,
      "learning_rate": 9.567220036464831e-05,
      "loss": 0.1501,
      "step": 1000
    },
    {
      "epoch": 0.91,
      "eval_loss": 0.1643093228340149,
      "eval_runtime": 49.3431,
      "eval_samples_per_second": 50.605,
      "eval_steps_per_second": 6.343,
      "step": 1000
    },
    {
      "epoch": 1.14,
      "learning_rate": 9.327319834948663e-05,
      "loss": 0.122,
      "step": 1250
    },
    {
      "epoch": 1.14,
      "eval_loss": 0.16473376750946045,
      "eval_runtime": 49.4257,
      "eval_samples_per_second": 50.52,
      "eval_steps_per_second": 6.333,
      "step": 1250
    },
    {
      "epoch": 1.37,
      "learning_rate": 9.087419633432492e-05,
      "loss": 0.1143,
      "step": 1500
    },
    {
      "epoch": 1.37,
      "eval_loss": 0.15297986567020416,
      "eval_runtime": 49.4063,
      "eval_samples_per_second": 50.54,
      "eval_steps_per_second": 6.335,
      "step": 1500
    },
    {
      "epoch": 1.59,
      "learning_rate": 8.847519431916324e-05,
      "loss": 0.1121,
      "step": 1750
    },
    {
      "epoch": 1.59,
      "eval_loss": 0.16007670760154724,
      "eval_runtime": 49.3369,
      "eval_samples_per_second": 50.611,
      "eval_steps_per_second": 6.344,
      "step": 1750
    },
    {
      "epoch": 1.82,
      "learning_rate": 8.607619230400153e-05,
      "loss": 0.1078,
      "step": 2000
    },
    {
      "epoch": 1.82,
      "eval_loss": 0.14469072222709656,
      "eval_runtime": 49.3439,
      "eval_samples_per_second": 50.604,
      "eval_steps_per_second": 6.343,
      "step": 2000
    },
    {
      "epoch": 2.05,
      "learning_rate": 8.367719028883985e-05,
      "loss": 0.0954,
      "step": 2250
    },
    {
      "epoch": 2.05,
      "eval_loss": 0.14710724353790283,
      "eval_runtime": 49.3536,
      "eval_samples_per_second": 50.594,
      "eval_steps_per_second": 6.342,
      "step": 2250
    },
    {
      "epoch": 2.28,
      "learning_rate": 8.127818827367816e-05,
      "loss": 0.0804,
      "step": 2500
    },
    {
      "epoch": 2.28,
      "eval_loss": 0.13685058057308197,
      "eval_runtime": 49.355,
      "eval_samples_per_second": 50.593,
      "eval_steps_per_second": 6.342,
      "step": 2500
    },
    {
      "epoch": 2.51,
      "learning_rate": 7.887918625851645e-05,
      "loss": 0.08,
      "step": 2750
    },
    {
      "epoch": 2.51,
      "eval_loss": 0.13740722835063934,
      "eval_runtime": 49.3607,
      "eval_samples_per_second": 50.587,
      "eval_steps_per_second": 6.341,
      "step": 2750
    },
    {
      "epoch": 2.73,
      "learning_rate": 7.648018424335477e-05,
      "loss": 0.0769,
      "step": 3000
    },
    {
      "epoch": 2.73,
      "eval_loss": 0.1370161473751068,
      "eval_runtime": 49.3701,
      "eval_samples_per_second": 50.577,
      "eval_steps_per_second": 6.34,
      "step": 3000
    },
    {
      "epoch": 2.96,
      "learning_rate": 7.408118222819308e-05,
      "loss": 0.0782,
      "step": 3250
    },
    {
      "epoch": 2.96,
      "eval_loss": 0.13374929130077362,
      "eval_runtime": 49.4074,
      "eval_samples_per_second": 50.539,
      "eval_steps_per_second": 6.335,
      "step": 3250
    },
    {
      "epoch": 3.19,
      "learning_rate": 7.168218021303138e-05,
      "loss": 0.0591,
      "step": 3500
    },
    {
      "epoch": 3.19,
      "eval_loss": 0.13687731325626373,
      "eval_runtime": 49.344,
      "eval_samples_per_second": 50.604,
      "eval_steps_per_second": 6.343,
      "step": 3500
    },
    {
      "epoch": 3.42,
      "learning_rate": 6.928317819786969e-05,
      "loss": 0.0575,
      "step": 3750
    },
    {
      "epoch": 3.42,
      "eval_loss": 0.13442662358283997,
      "eval_runtime": 49.347,
      "eval_samples_per_second": 50.601,
      "eval_steps_per_second": 6.343,
      "step": 3750
    },
    {
      "epoch": 3.64,
      "learning_rate": 6.6884176182708e-05,
      "loss": 0.0579,
      "step": 4000
    },
    {
      "epoch": 3.64,
      "eval_loss": 0.13463211059570312,
      "eval_runtime": 49.3532,
      "eval_samples_per_second": 50.595,
      "eval_steps_per_second": 6.342,
      "step": 4000
    },
    {
      "epoch": 3.87,
      "learning_rate": 6.44851741675463e-05,
      "loss": 0.0541,
      "step": 4250
    },
    {
      "epoch": 3.87,
      "eval_loss": 0.12762367725372314,
      "eval_runtime": 49.3131,
      "eval_samples_per_second": 50.636,
      "eval_steps_per_second": 6.347,
      "step": 4250
    },
    {
      "epoch": 4.1,
      "learning_rate": 6.208617215238462e-05,
      "loss": 0.0469,
      "step": 4500
    },
    {
      "epoch": 4.1,
      "eval_loss": 0.1321054846048355,
      "eval_runtime": 49.3382,
      "eval_samples_per_second": 50.61,
      "eval_steps_per_second": 6.344,
      "step": 4500
    },
    {
      "epoch": 4.33,
      "learning_rate": 5.968717013722291e-05,
      "loss": 0.0367,
      "step": 4750
    },
    {
      "epoch": 4.33,
      "eval_loss": 0.13329076766967773,
      "eval_runtime": 49.3411,
      "eval_samples_per_second": 50.607,
      "eval_steps_per_second": 6.344,
      "step": 4750
    },
    {
      "epoch": 4.55,
      "learning_rate": 5.7288168122061226e-05,
      "loss": 0.0409,
      "step": 5000
    },
    {
      "epoch": 4.55,
      "eval_loss": 0.13458645343780518,
      "eval_runtime": 49.3393,
      "eval_samples_per_second": 50.609,
      "eval_steps_per_second": 6.344,
      "step": 5000
    },
    {
      "epoch": 4.78,
      "learning_rate": 5.488916610689954e-05,
      "loss": 0.0402,
      "step": 5250
    },
    {
      "epoch": 4.78,
      "eval_loss": 0.12923233211040497,
      "eval_runtime": 49.4868,
      "eval_samples_per_second": 50.458,
      "eval_steps_per_second": 6.325,
      "step": 5250
    },
    {
      "epoch": 5.01,
      "learning_rate": 5.249016409173784e-05,
      "loss": 0.0378,
      "step": 5500
    },
    {
      "epoch": 5.01,
      "eval_loss": 0.12460647523403168,
      "eval_runtime": 49.3719,
      "eval_samples_per_second": 50.575,
      "eval_steps_per_second": 6.34,
      "step": 5500
    },
    {
      "epoch": 5.24,
      "learning_rate": 5.009116207657615e-05,
      "loss": 0.0258,
      "step": 5750
    },
    {
      "epoch": 5.24,
      "eval_loss": 0.1305789053440094,
      "eval_runtime": 49.3929,
      "eval_samples_per_second": 50.554,
      "eval_steps_per_second": 6.337,
      "step": 5750
    },
    {
      "epoch": 5.47,
      "learning_rate": 4.769216006141446e-05,
      "loss": 0.0252,
      "step": 6000
    },
    {
      "epoch": 5.47,
      "eval_loss": 0.13075487315654755,
      "eval_runtime": 49.3793,
      "eval_samples_per_second": 50.568,
      "eval_steps_per_second": 6.339,
      "step": 6000
    },
    {
      "epoch": 5.69,
      "learning_rate": 4.5293158046252756e-05,
      "loss": 0.0266,
      "step": 6250
    },
    {
      "epoch": 5.69,
      "eval_loss": 0.13152019679546356,
      "eval_runtime": 49.3918,
      "eval_samples_per_second": 50.555,
      "eval_steps_per_second": 6.337,
      "step": 6250
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.289415603109107e-05,
      "loss": 0.0264,
      "step": 6500
    },
    {
      "epoch": 5.92,
      "eval_loss": 0.12978705763816833,
      "eval_runtime": 49.4158,
      "eval_samples_per_second": 50.53,
      "eval_steps_per_second": 6.334,
      "step": 6500
    },
    {
      "epoch": 6.15,
      "learning_rate": 4.0495154015929375e-05,
      "loss": 0.0204,
      "step": 6750
    },
    {
      "epoch": 6.15,
      "eval_loss": 0.1330789029598236,
      "eval_runtime": 49.4096,
      "eval_samples_per_second": 50.537,
      "eval_steps_per_second": 6.335,
      "step": 6750
    },
    {
      "epoch": 6.38,
      "learning_rate": 3.809615200076768e-05,
      "loss": 0.0176,
      "step": 7000
    },
    {
      "epoch": 6.38,
      "eval_loss": 0.13327623903751373,
      "eval_runtime": 49.4281,
      "eval_samples_per_second": 50.518,
      "eval_steps_per_second": 6.332,
      "step": 7000
    },
    {
      "epoch": 6.6,
      "learning_rate": 3.569714998560599e-05,
      "loss": 0.0177,
      "step": 7250
    },
    {
      "epoch": 6.6,
      "eval_loss": 0.13123974204063416,
      "eval_runtime": 49.4644,
      "eval_samples_per_second": 50.481,
      "eval_steps_per_second": 6.328,
      "step": 7250
    },
    {
      "epoch": 6.83,
      "learning_rate": 3.32981479704443e-05,
      "loss": 0.0161,
      "step": 7500
    },
    {
      "epoch": 6.83,
      "eval_loss": 0.1328614354133606,
      "eval_runtime": 49.4499,
      "eval_samples_per_second": 50.496,
      "eval_steps_per_second": 6.33,
      "step": 7500
    },
    {
      "epoch": 7.06,
      "learning_rate": 3.0899145955282606e-05,
      "loss": 0.016,
      "step": 7750
    },
    {
      "epoch": 7.06,
      "eval_loss": 0.13026753067970276,
      "eval_runtime": 49.4661,
      "eval_samples_per_second": 50.479,
      "eval_steps_per_second": 6.328,
      "step": 7750
    },
    {
      "epoch": 7.29,
      "learning_rate": 2.850014394012091e-05,
      "loss": 0.0104,
      "step": 8000
    },
    {
      "epoch": 7.29,
      "eval_loss": 0.13250969350337982,
      "eval_runtime": 49.5109,
      "eval_samples_per_second": 50.433,
      "eval_steps_per_second": 6.322,
      "step": 8000
    },
    {
      "epoch": 7.52,
      "learning_rate": 2.6101141924959215e-05,
      "loss": 0.0104,
      "step": 8250
    },
    {
      "epoch": 7.52,
      "eval_loss": 0.1344473958015442,
      "eval_runtime": 49.4545,
      "eval_samples_per_second": 50.491,
      "eval_steps_per_second": 6.329,
      "step": 8250
    },
    {
      "epoch": 7.74,
      "learning_rate": 2.3702139909797524e-05,
      "loss": 0.0107,
      "step": 8500
    },
    {
      "epoch": 7.74,
      "eval_loss": 0.13361412286758423,
      "eval_runtime": 49.4611,
      "eval_samples_per_second": 50.484,
      "eval_steps_per_second": 6.328,
      "step": 8500
    },
    {
      "epoch": 7.97,
      "learning_rate": 2.1303137894635834e-05,
      "loss": 0.0105,
      "step": 8750
    },
    {
      "epoch": 7.97,
      "eval_loss": 0.1311049610376358,
      "eval_runtime": 49.3899,
      "eval_samples_per_second": 50.557,
      "eval_steps_per_second": 6.337,
      "step": 8750
    },
    {
      "epoch": 8.2,
      "learning_rate": 1.890413587947414e-05,
      "loss": 0.0072,
      "step": 9000
    },
    {
      "epoch": 8.2,
      "eval_loss": 0.1345677375793457,
      "eval_runtime": 49.4945,
      "eval_samples_per_second": 50.45,
      "eval_steps_per_second": 6.324,
      "step": 9000
    },
    {
      "epoch": 8.43,
      "learning_rate": 1.6505133864312446e-05,
      "loss": 0.0065,
      "step": 9250
    },
    {
      "epoch": 8.43,
      "eval_loss": 0.13423801958560944,
      "eval_runtime": 49.4363,
      "eval_samples_per_second": 50.509,
      "eval_steps_per_second": 6.331,
      "step": 9250
    },
    {
      "epoch": 8.65,
      "learning_rate": 1.4106131849150753e-05,
      "loss": 0.0062,
      "step": 9500
    },
    {
      "epoch": 8.65,
      "eval_loss": 0.13279776275157928,
      "eval_runtime": 49.5198,
      "eval_samples_per_second": 50.424,
      "eval_steps_per_second": 6.321,
      "step": 9500
    },
    {
      "epoch": 8.88,
      "learning_rate": 1.1707129833989061e-05,
      "loss": 0.006,
      "step": 9750
    },
    {
      "epoch": 8.88,
      "eval_loss": 0.13258913159370422,
      "eval_runtime": 49.511,
      "eval_samples_per_second": 50.433,
      "eval_steps_per_second": 6.322,
      "step": 9750
    },
    {
      "epoch": 9.11,
      "learning_rate": 9.308127818827369e-06,
      "loss": 0.0052,
      "step": 10000
    },
    {
      "epoch": 9.11,
      "eval_loss": 0.13228829205036163,
      "eval_runtime": 49.4456,
      "eval_samples_per_second": 50.5,
      "eval_steps_per_second": 6.33,
      "step": 10000
    },
    {
      "epoch": 9.34,
      "learning_rate": 6.909125803665675e-06,
      "loss": 0.0039,
      "step": 10250
    },
    {
      "epoch": 9.34,
      "eval_loss": 0.13294672966003418,
      "eval_runtime": 49.4619,
      "eval_samples_per_second": 50.483,
      "eval_steps_per_second": 6.328,
      "step": 10250
    },
    {
      "epoch": 9.57,
      "learning_rate": 4.510123788503983e-06,
      "loss": 0.0039,
      "step": 10500
    },
    {
      "epoch": 9.57,
      "eval_loss": 0.13272705674171448,
      "eval_runtime": 49.4568,
      "eval_samples_per_second": 50.489,
      "eval_steps_per_second": 6.329,
      "step": 10500
    },
    {
      "epoch": 9.79,
      "learning_rate": 2.11112177334229e-06,
      "loss": 0.004,
      "step": 10750
    },
    {
      "epoch": 9.79,
      "eval_loss": 0.13220719993114471,
      "eval_runtime": 49.465,
      "eval_samples_per_second": 50.48,
      "eval_steps_per_second": 6.328,
      "step": 10750
    }
  ],
  "max_steps": 10970,
  "num_train_epochs": 10,
  "total_flos": 8.710540889772442e+16,
  "trial_name": null,
  "trial_params": null
}