{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.47292732361741147,
  "eval_steps": 100,
  "global_step": 4900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "grad_norm": 5.391254425048828,
      "learning_rate": 2.450490098019604e-05,
      "loss": 1.8358,
      "step": 100
    },
    {
      "epoch": 0.01,
      "eval_loss": 1.450282096862793,
      "eval_runtime": 27.2888,
      "eval_samples_per_second": 18.323,
      "eval_steps_per_second": 2.309,
      "step": 100
    },
    {
      "epoch": 0.02,
      "grad_norm": 5.370517253875732,
      "learning_rate": 2.400480096019204e-05,
      "loss": 1.6054,
      "step": 200
    },
    {
      "epoch": 0.02,
      "eval_loss": 1.4298174381256104,
      "eval_runtime": 27.2786,
      "eval_samples_per_second": 18.329,
      "eval_steps_per_second": 2.31,
      "step": 200
    },
    {
      "epoch": 0.03,
      "grad_norm": 5.93058443069458,
      "learning_rate": 2.3504700940188037e-05,
      "loss": 1.7294,
      "step": 300
    },
    {
      "epoch": 0.03,
      "eval_loss": 1.404078483581543,
      "eval_runtime": 27.2841,
      "eval_samples_per_second": 18.326,
      "eval_steps_per_second": 2.309,
      "step": 300
    },
    {
      "epoch": 0.04,
      "grad_norm": 3.4010226726531982,
      "learning_rate": 2.3004600920184037e-05,
      "loss": 1.591,
      "step": 400
    },
    {
      "epoch": 0.04,
      "eval_loss": 1.3933534622192383,
      "eval_runtime": 27.2614,
      "eval_samples_per_second": 18.341,
      "eval_steps_per_second": 2.311,
      "step": 400
    },
    {
      "epoch": 0.05,
      "grad_norm": 5.741560459136963,
      "learning_rate": 2.2504500900180037e-05,
      "loss": 1.6501,
      "step": 500
    },
    {
      "epoch": 0.05,
      "eval_loss": 1.3843183517456055,
      "eval_runtime": 27.2728,
      "eval_samples_per_second": 18.333,
      "eval_steps_per_second": 2.31,
      "step": 500
    },
    {
      "epoch": 0.06,
      "grad_norm": 3.929501533508301,
      "learning_rate": 2.2004400880176038e-05,
      "loss": 1.5758,
      "step": 600
    },
    {
      "epoch": 0.06,
      "eval_loss": 1.3718712329864502,
      "eval_runtime": 27.2551,
      "eval_samples_per_second": 18.345,
      "eval_steps_per_second": 2.311,
      "step": 600
    },
    {
      "epoch": 0.07,
      "grad_norm": 6.033923149108887,
      "learning_rate": 2.1504300860172034e-05,
      "loss": 1.6159,
      "step": 700
    },
    {
      "epoch": 0.07,
      "eval_loss": 1.377557396888733,
      "eval_runtime": 27.2622,
      "eval_samples_per_second": 18.34,
      "eval_steps_per_second": 2.311,
      "step": 700
    },
    {
      "epoch": 0.08,
      "grad_norm": 7.7004313468933105,
      "learning_rate": 2.1004200840168034e-05,
      "loss": 1.6272,
      "step": 800
    },
    {
      "epoch": 0.08,
      "eval_loss": 1.3639216423034668,
      "eval_runtime": 27.271,
      "eval_samples_per_second": 18.334,
      "eval_steps_per_second": 2.31,
      "step": 800
    },
    {
      "epoch": 0.09,
      "grad_norm": 3.8936071395874023,
      "learning_rate": 2.0504100820164035e-05,
      "loss": 1.5669,
      "step": 900
    },
    {
      "epoch": 0.09,
      "eval_loss": 1.3606789112091064,
      "eval_runtime": 27.2614,
      "eval_samples_per_second": 18.341,
      "eval_steps_per_second": 2.311,
      "step": 900
    },
    {
      "epoch": 0.1,
      "grad_norm": 2.6616313457489014,
      "learning_rate": 2.0004000800160035e-05,
      "loss": 1.5046,
      "step": 1000
    },
    {
      "epoch": 0.1,
      "eval_loss": 1.3455876111984253,
      "eval_runtime": 27.2532,
      "eval_samples_per_second": 18.346,
      "eval_steps_per_second": 2.312,
      "step": 1000
    },
    {
      "epoch": 0.11,
      "grad_norm": 4.210849761962891,
      "learning_rate": 1.950390078015603e-05,
      "loss": 1.5338,
      "step": 1100
    },
    {
      "epoch": 0.11,
      "eval_loss": 1.3470462560653687,
      "eval_runtime": 27.2373,
      "eval_samples_per_second": 18.357,
      "eval_steps_per_second": 2.313,
      "step": 1100
    },
    {
      "epoch": 0.12,
      "grad_norm": 7.99553108215332,
      "learning_rate": 1.900380076015203e-05,
      "loss": 1.5559,
      "step": 1200
    },
    {
      "epoch": 0.12,
      "eval_loss": 1.3312621116638184,
      "eval_runtime": 27.2565,
      "eval_samples_per_second": 18.344,
      "eval_steps_per_second": 2.311,
      "step": 1200
    },
    {
      "epoch": 0.13,
      "grad_norm": 7.216369152069092,
      "learning_rate": 1.8503700740148032e-05,
      "loss": 1.5616,
      "step": 1300
    },
    {
      "epoch": 0.13,
      "eval_loss": 1.3286023139953613,
      "eval_runtime": 27.2461,
      "eval_samples_per_second": 18.351,
      "eval_steps_per_second": 2.312,
      "step": 1300
    },
    {
      "epoch": 0.14,
      "grad_norm": 3.162696361541748,
      "learning_rate": 1.800360072014403e-05,
      "loss": 1.4969,
      "step": 1400
    },
    {
      "epoch": 0.14,
      "eval_loss": 1.3286837339401245,
      "eval_runtime": 27.2552,
      "eval_samples_per_second": 18.345,
      "eval_steps_per_second": 2.311,
      "step": 1400
    },
    {
      "epoch": 0.14,
      "grad_norm": 7.42529821395874,
      "learning_rate": 1.750350070014003e-05,
      "loss": 1.5723,
      "step": 1500
    },
    {
      "epoch": 0.14,
      "eval_loss": 1.3215287923812866,
      "eval_runtime": 27.2475,
      "eval_samples_per_second": 18.35,
      "eval_steps_per_second": 2.312,
      "step": 1500
    },
    {
      "epoch": 0.15,
      "grad_norm": 2.7483532428741455,
      "learning_rate": 1.700340068013603e-05,
      "loss": 1.4844,
      "step": 1600
    },
    {
      "epoch": 0.15,
      "eval_loss": 1.3189783096313477,
      "eval_runtime": 27.2412,
      "eval_samples_per_second": 18.355,
      "eval_steps_per_second": 2.313,
      "step": 1600
    },
    {
      "epoch": 0.16,
      "grad_norm": 11.681495666503906,
      "learning_rate": 1.6503300660132026e-05,
      "loss": 1.5056,
      "step": 1700
    },
    {
      "epoch": 0.16,
      "eval_loss": 1.313588261604309,
      "eval_runtime": 27.2711,
      "eval_samples_per_second": 18.334,
      "eval_steps_per_second": 2.31,
      "step": 1700
    },
    {
      "epoch": 0.17,
      "grad_norm": 6.721160888671875,
      "learning_rate": 1.6003200640128026e-05,
      "loss": 1.5118,
      "step": 1800
    },
    {
      "epoch": 0.17,
      "eval_loss": 1.3094464540481567,
      "eval_runtime": 27.2681,
      "eval_samples_per_second": 18.336,
      "eval_steps_per_second": 2.31,
      "step": 1800
    },
    {
      "epoch": 0.18,
      "grad_norm": 6.978147029876709,
      "learning_rate": 1.5503100620124026e-05,
      "loss": 1.4269,
      "step": 1900
    },
    {
      "epoch": 0.18,
      "eval_loss": 1.3047374486923218,
      "eval_runtime": 27.289,
      "eval_samples_per_second": 18.322,
      "eval_steps_per_second": 2.309,
      "step": 1900
    },
    {
      "epoch": 0.19,
      "grad_norm": 6.569401741027832,
      "learning_rate": 1.5003000600120023e-05,
      "loss": 1.6078,
      "step": 2000
    },
    {
      "epoch": 0.19,
      "eval_loss": 1.2957228422164917,
      "eval_runtime": 27.2723,
      "eval_samples_per_second": 18.334,
      "eval_steps_per_second": 2.31,
      "step": 2000
    },
    {
      "epoch": 0.2,
      "grad_norm": 7.020974159240723,
      "learning_rate": 1.4502900580116025e-05,
      "loss": 1.5223,
      "step": 2100
    },
    {
      "epoch": 0.2,
      "eval_loss": 1.2962332963943481,
      "eval_runtime": 27.2629,
      "eval_samples_per_second": 18.34,
      "eval_steps_per_second": 2.311,
      "step": 2100
    },
    {
      "epoch": 0.21,
      "grad_norm": 11.144767761230469,
      "learning_rate": 1.4002800560112023e-05,
      "loss": 1.5419,
      "step": 2200
    },
    {
      "epoch": 0.21,
      "eval_loss": 1.2900757789611816,
      "eval_runtime": 27.2619,
      "eval_samples_per_second": 18.341,
      "eval_steps_per_second": 2.311,
      "step": 2200
    },
    {
      "epoch": 0.22,
      "grad_norm": 8.033313751220703,
      "learning_rate": 1.3502700540108021e-05,
      "loss": 1.5363,
      "step": 2300
    },
    {
      "epoch": 0.22,
      "eval_loss": 1.2853684425354004,
      "eval_runtime": 27.2673,
      "eval_samples_per_second": 18.337,
      "eval_steps_per_second": 2.31,
      "step": 2300
    },
    {
      "epoch": 0.23,
      "grad_norm": 8.654736518859863,
      "learning_rate": 1.3002600520104022e-05,
      "loss": 1.4545,
      "step": 2400
    },
    {
      "epoch": 0.23,
      "eval_loss": 1.2811810970306396,
      "eval_runtime": 27.2594,
      "eval_samples_per_second": 18.342,
      "eval_steps_per_second": 2.311,
      "step": 2400
    },
    {
      "epoch": 0.24,
      "grad_norm": 9.393715858459473,
      "learning_rate": 1.250250050010002e-05,
      "loss": 1.4054,
      "step": 2500
    },
    {
      "epoch": 0.24,
      "eval_loss": 1.2781436443328857,
      "eval_runtime": 27.2437,
      "eval_samples_per_second": 18.353,
      "eval_steps_per_second": 2.312,
      "step": 2500
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.9280712604522705,
      "learning_rate": 1.200240048009602e-05,
      "loss": 1.5613,
      "step": 2600
    },
    {
      "epoch": 0.25,
      "eval_loss": 1.2793227434158325,
      "eval_runtime": 27.2593,
      "eval_samples_per_second": 18.342,
      "eval_steps_per_second": 2.311,
      "step": 2600
    },
    {
      "epoch": 0.26,
      "grad_norm": 9.87479019165039,
      "learning_rate": 1.1502300460092019e-05,
      "loss": 1.3906,
      "step": 2700
    },
    {
      "epoch": 0.26,
      "eval_loss": 1.2742949724197388,
      "eval_runtime": 27.2694,
      "eval_samples_per_second": 18.336,
      "eval_steps_per_second": 2.31,
      "step": 2700
    },
    {
      "epoch": 0.27,
      "grad_norm": 8.556775093078613,
      "learning_rate": 1.1002200440088019e-05,
      "loss": 1.5082,
      "step": 2800
    },
    {
      "epoch": 0.27,
      "eval_loss": 1.2731441259384155,
      "eval_runtime": 27.2537,
      "eval_samples_per_second": 18.346,
      "eval_steps_per_second": 2.312,
      "step": 2800
    },
    {
      "epoch": 0.28,
      "grad_norm": 4.113871097564697,
      "learning_rate": 1.0502100420084017e-05,
      "loss": 1.4439,
      "step": 2900
    },
    {
      "epoch": 0.28,
      "eval_loss": 1.2708263397216797,
      "eval_runtime": 27.2642,
      "eval_samples_per_second": 18.339,
      "eval_steps_per_second": 2.311,
      "step": 2900
    },
    {
      "epoch": 0.29,
      "grad_norm": 7.74875545501709,
      "learning_rate": 1.0002000400080017e-05,
      "loss": 1.4189,
      "step": 3000
    },
    {
      "epoch": 0.29,
      "eval_loss": 1.2717357873916626,
      "eval_runtime": 27.2628,
      "eval_samples_per_second": 18.34,
      "eval_steps_per_second": 2.311,
      "step": 3000
    },
    {
      "epoch": 0.3,
      "grad_norm": 3.830393075942993,
      "learning_rate": 9.501900380076016e-06,
      "loss": 1.4567,
      "step": 3100
    },
    {
      "epoch": 0.3,
      "eval_loss": 1.2686272859573364,
      "eval_runtime": 27.2577,
      "eval_samples_per_second": 18.343,
      "eval_steps_per_second": 2.311,
      "step": 3100
    },
    {
      "epoch": 0.31,
      "grad_norm": 7.333063125610352,
      "learning_rate": 9.001800360072014e-06,
      "loss": 1.4613,
      "step": 3200
    },
    {
      "epoch": 0.31,
      "eval_loss": 1.266152262687683,
      "eval_runtime": 27.2615,
      "eval_samples_per_second": 18.341,
      "eval_steps_per_second": 2.311,
      "step": 3200
    },
    {
      "epoch": 0.32,
      "grad_norm": 2.688906192779541,
      "learning_rate": 8.501700340068014e-06,
      "loss": 1.4106,
      "step": 3300
    },
    {
      "epoch": 0.32,
      "eval_loss": 1.2621879577636719,
      "eval_runtime": 27.2709,
      "eval_samples_per_second": 18.335,
      "eval_steps_per_second": 2.31,
      "step": 3300
    },
    {
      "epoch": 0.33,
      "grad_norm": 5.106052398681641,
      "learning_rate": 8.001600320064013e-06,
      "loss": 1.4848,
      "step": 3400
    },
    {
      "epoch": 0.33,
      "eval_loss": 1.259779453277588,
      "eval_runtime": 27.2724,
      "eval_samples_per_second": 18.334,
      "eval_steps_per_second": 2.31,
      "step": 3400
    },
    {
      "epoch": 0.34,
      "grad_norm": 7.964311599731445,
      "learning_rate": 7.501500300060011e-06,
      "loss": 1.3907,
      "step": 3500
    },
    {
      "epoch": 0.34,
      "eval_loss": 1.2596435546875,
      "eval_runtime": 27.2626,
      "eval_samples_per_second": 18.34,
      "eval_steps_per_second": 2.311,
      "step": 3500
    },
    {
      "epoch": 0.35,
      "grad_norm": 6.268681526184082,
      "learning_rate": 7.0014002800560115e-06,
      "loss": 1.44,
      "step": 3600
    },
    {
      "epoch": 0.35,
      "eval_loss": 1.2597826719284058,
      "eval_runtime": 27.2599,
      "eval_samples_per_second": 18.342,
      "eval_steps_per_second": 2.311,
      "step": 3600
    },
    {
      "epoch": 0.36,
      "grad_norm": 3.744537115097046,
      "learning_rate": 6.501300260052011e-06,
      "loss": 1.4985,
      "step": 3700
    },
    {
      "epoch": 0.36,
      "eval_loss": 1.2589830160140991,
      "eval_runtime": 27.2604,
      "eval_samples_per_second": 18.342,
      "eval_steps_per_second": 2.311,
      "step": 3700
    },
    {
      "epoch": 0.37,
      "grad_norm": 6.038290500640869,
      "learning_rate": 6.00120024004801e-06,
      "loss": 1.3521,
      "step": 3800
    },
    {
      "epoch": 0.37,
      "eval_loss": 1.255460500717163,
      "eval_runtime": 27.2617,
      "eval_samples_per_second": 18.341,
      "eval_steps_per_second": 2.311,
      "step": 3800
    },
    {
      "epoch": 0.38,
      "grad_norm": 8.274654388427734,
      "learning_rate": 5.501100220044009e-06,
      "loss": 1.4754,
      "step": 3900
    },
    {
      "epoch": 0.38,
      "eval_loss": 1.254140019416809,
      "eval_runtime": 27.2664,
      "eval_samples_per_second": 18.338,
      "eval_steps_per_second": 2.311,
      "step": 3900
    },
    {
      "epoch": 0.39,
      "grad_norm": 6.011349201202393,
      "learning_rate": 5.001000200040009e-06,
      "loss": 1.3927,
      "step": 4000
    },
    {
      "epoch": 0.39,
      "eval_loss": 1.252474069595337,
      "eval_runtime": 27.2815,
      "eval_samples_per_second": 18.327,
      "eval_steps_per_second": 2.309,
      "step": 4000
    },
    {
      "epoch": 0.4,
      "grad_norm": 8.328156471252441,
      "learning_rate": 4.500900180036007e-06,
      "loss": 1.468,
      "step": 4100
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.2514727115631104,
      "eval_runtime": 27.2652,
      "eval_samples_per_second": 18.338,
      "eval_steps_per_second": 2.311,
      "step": 4100
    },
    {
      "epoch": 0.41,
      "grad_norm": 7.536208152770996,
      "learning_rate": 4.0008001600320064e-06,
      "loss": 1.4471,
      "step": 4200
    },
    {
      "epoch": 0.41,
      "eval_loss": 1.2502838373184204,
      "eval_runtime": 27.2698,
      "eval_samples_per_second": 18.335,
      "eval_steps_per_second": 2.31,
      "step": 4200
    },
    {
      "epoch": 0.42,
      "grad_norm": 6.996093273162842,
      "learning_rate": 3.5007001400280057e-06,
      "loss": 1.445,
      "step": 4300
    },
    {
      "epoch": 0.42,
      "eval_loss": 1.2483798265457153,
      "eval_runtime": 27.2509,
      "eval_samples_per_second": 18.348,
      "eval_steps_per_second": 2.312,
      "step": 4300
    },
    {
      "epoch": 0.42,
      "grad_norm": 5.2884063720703125,
      "learning_rate": 3.000600120024005e-06,
      "loss": 1.465,
      "step": 4400
    },
    {
      "epoch": 0.42,
      "eval_loss": 1.2479716539382935,
      "eval_runtime": 27.2406,
      "eval_samples_per_second": 18.355,
      "eval_steps_per_second": 2.313,
      "step": 4400
    },
    {
      "epoch": 0.43,
      "grad_norm": 6.742663383483887,
      "learning_rate": 2.5005001000200043e-06,
      "loss": 1.3964,
      "step": 4500
    },
    {
      "epoch": 0.43,
      "eval_loss": 1.2463291883468628,
      "eval_runtime": 27.2445,
      "eval_samples_per_second": 18.352,
      "eval_steps_per_second": 2.312,
      "step": 4500
    },
    {
      "epoch": 0.44,
      "grad_norm": 5.316954612731934,
      "learning_rate": 2.0004000800160032e-06,
      "loss": 1.4355,
      "step": 4600
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.244441032409668,
      "eval_runtime": 27.2539,
      "eval_samples_per_second": 18.346,
      "eval_steps_per_second": 2.312,
      "step": 4600
    },
    {
      "epoch": 0.45,
      "grad_norm": 7.313477516174316,
      "learning_rate": 1.5003000600120025e-06,
      "loss": 1.4902,
      "step": 4700
    },
    {
      "epoch": 0.45,
      "eval_loss": 1.2442010641098022,
      "eval_runtime": 27.2614,
      "eval_samples_per_second": 18.341,
      "eval_steps_per_second": 2.311,
      "step": 4700
    },
    {
      "epoch": 0.46,
      "grad_norm": 4.464217185974121,
      "learning_rate": 1.0002000400080016e-06,
      "loss": 1.3817,
      "step": 4800
    },
    {
      "epoch": 0.46,
      "eval_loss": 1.2431871891021729,
      "eval_runtime": 27.2574,
      "eval_samples_per_second": 18.344,
      "eval_steps_per_second": 2.311,
      "step": 4800
    },
    {
      "epoch": 0.47,
      "grad_norm": 9.320337295532227,
      "learning_rate": 5.001000200040008e-07,
      "loss": 1.3901,
      "step": 4900
    },
    {
      "epoch": 0.47,
      "eval_loss": 1.2422610521316528,
      "eval_runtime": 27.2557,
      "eval_samples_per_second": 18.345,
      "eval_steps_per_second": 2.311,
      "step": 4900
    }
  ],
  "logging_steps": 100,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "total_flos": 1.866856780959744e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}