{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 6.524030177821643,
  "eval_steps": 2500,
  "global_step": 97500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03,
      "learning_rate": 0.0005,
      "loss": 0.5631,
      "step": 500
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0004974874371859296,
      "loss": 0.14,
      "step": 1000
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0004949748743718593,
      "loss": 0.0804,
      "step": 1500
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.0004924623115577889,
      "loss": 0.0597,
      "step": 2000
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0004899497487437187,
      "loss": 0.0501,
      "step": 2500
    },
    {
      "epoch": 0.17,
      "eval_loss": 0.033905673772096634,
      "eval_runtime": 8.0076,
      "eval_samples_per_second": 124.881,
      "eval_steps_per_second": 1.249,
      "step": 2500
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00048743718592964827,
      "loss": 0.0436,
      "step": 3000
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0004849246231155779,
      "loss": 0.0398,
      "step": 3500
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00048241206030150754,
      "loss": 0.0374,
      "step": 4000
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0004798994974874372,
      "loss": 0.035,
      "step": 4500
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00047738693467336686,
      "loss": 0.0331,
      "step": 5000
    },
    {
      "epoch": 0.33,
      "eval_loss": 0.023899264633655548,
      "eval_runtime": 7.9042,
      "eval_samples_per_second": 126.515,
      "eval_steps_per_second": 1.265,
      "step": 5000
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00047487437185929647,
      "loss": 0.0313,
      "step": 5500
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00047236180904522613,
      "loss": 0.03,
      "step": 6000
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0004698492462311558,
      "loss": 0.0287,
      "step": 6500
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00046733668341708545,
      "loss": 0.0277,
      "step": 7000
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0004648241206030151,
      "loss": 0.0271,
      "step": 7500
    },
    {
      "epoch": 0.5,
      "eval_loss": 0.01935673877596855,
      "eval_runtime": 8.4605,
      "eval_samples_per_second": 118.197,
      "eval_steps_per_second": 1.182,
      "step": 7500
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0004623115577889447,
      "loss": 0.0262,
      "step": 8000
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0004597989949748744,
      "loss": 0.0253,
      "step": 8500
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.000457286432160804,
      "loss": 0.0244,
      "step": 9000
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0004547738693467337,
      "loss": 0.0239,
      "step": 9500
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0004522613065326633,
      "loss": 0.0238,
      "step": 10000
    },
    {
      "epoch": 0.67,
      "eval_loss": 0.017778538167476654,
      "eval_runtime": 7.6198,
      "eval_samples_per_second": 131.237,
      "eval_steps_per_second": 1.312,
      "step": 10000
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.000449748743718593,
      "loss": 0.0229,
      "step": 10500
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0004472361809045226,
      "loss": 0.0223,
      "step": 11000
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00044472361809045225,
      "loss": 0.0219,
      "step": 11500
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00044221105527638197,
      "loss": 0.0216,
      "step": 12000
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0004396984924623116,
      "loss": 0.0212,
      "step": 12500
    },
    {
      "epoch": 0.84,
      "eval_loss": 0.014742564409971237,
      "eval_runtime": 7.6159,
      "eval_samples_per_second": 131.305,
      "eval_steps_per_second": 1.313,
      "step": 12500
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00043718592964824124,
      "loss": 0.0206,
      "step": 13000
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00043467336683417085,
      "loss": 0.0205,
      "step": 13500
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0004321608040201005,
      "loss": 0.0198,
      "step": 14000
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0004296482412060301,
      "loss": 0.0198,
      "step": 14500
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00042713567839195983,
      "loss": 0.0193,
      "step": 15000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.01463437918573618,
      "eval_runtime": 7.6369,
      "eval_samples_per_second": 130.944,
      "eval_steps_per_second": 1.309,
      "step": 15000
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00042462311557788944,
      "loss": 0.0178,
      "step": 15500
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0004221105527638191,
      "loss": 0.0177,
      "step": 16000
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00041959798994974876,
      "loss": 0.0172,
      "step": 16500
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00041708542713567837,
      "loss": 0.017,
      "step": 17000
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0004145728643216081,
      "loss": 0.0172,
      "step": 17500
    },
    {
      "epoch": 1.17,
      "eval_loss": 0.0131097212433815,
      "eval_runtime": 7.63,
      "eval_samples_per_second": 131.062,
      "eval_steps_per_second": 1.311,
      "step": 17500
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0004120603015075377,
      "loss": 0.0168,
      "step": 18000
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00040954773869346736,
      "loss": 0.017,
      "step": 18500
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00040703517587939697,
      "loss": 0.0165,
      "step": 19000
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00040452261306532663,
      "loss": 0.0163,
      "step": 19500
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0004020100502512563,
      "loss": 0.0163,
      "step": 20000
    },
    {
      "epoch": 1.34,
      "eval_loss": 0.012624073773622513,
      "eval_runtime": 7.6613,
      "eval_samples_per_second": 130.527,
      "eval_steps_per_second": 1.305,
      "step": 20000
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00039949748743718595,
      "loss": 0.0161,
      "step": 20500
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00039698492462311556,
      "loss": 0.0162,
      "step": 21000
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0003944723618090452,
      "loss": 0.0162,
      "step": 21500
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.0003919597989949749,
      "loss": 0.0158,
      "step": 22000
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00038944723618090455,
      "loss": 0.0157,
      "step": 22500
    },
    {
      "epoch": 1.51,
      "eval_loss": 0.0122187789529562,
      "eval_runtime": 7.6253,
      "eval_samples_per_second": 131.143,
      "eval_steps_per_second": 1.311,
      "step": 22500
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0003869346733668342,
      "loss": 0.0289,
      "step": 23000
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0003844221105527638,
      "loss": 0.0157,
      "step": 23500
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0003819095477386935,
      "loss": 0.0148,
      "step": 24000
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0003793969849246231,
      "loss": 0.0148,
      "step": 24500
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0003768844221105528,
      "loss": 0.0147,
      "step": 25000
    },
    {
      "epoch": 1.67,
      "eval_loss": 0.011391976848244667,
      "eval_runtime": 7.5953,
      "eval_samples_per_second": 131.66,
      "eval_steps_per_second": 1.317,
      "step": 25000
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.0003743718592964824,
      "loss": 0.0148,
      "step": 25500
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00037185929648241207,
      "loss": 0.0146,
      "step": 26000
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0003693467336683417,
      "loss": 0.0145,
      "step": 26500
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00036683417085427134,
      "loss": 0.0143,
      "step": 27000
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00036432160804020106,
      "loss": 0.0144,
      "step": 27500
    },
    {
      "epoch": 1.84,
      "eval_loss": 0.011031309142708778,
      "eval_runtime": 7.6444,
      "eval_samples_per_second": 130.814,
      "eval_steps_per_second": 1.308,
      "step": 27500
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00036180904522613067,
      "loss": 0.0142,
      "step": 28000
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00035929648241206033,
      "loss": 0.014,
      "step": 28500
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00035678391959798994,
      "loss": 0.0143,
      "step": 29000
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0003542713567839196,
      "loss": 0.0144,
      "step": 29500
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.00035175879396984926,
      "loss": 0.0136,
      "step": 30000
    },
    {
      "epoch": 2.01,
      "eval_loss": 0.010939952917397022,
      "eval_runtime": 7.6117,
      "eval_samples_per_second": 131.376,
      "eval_steps_per_second": 1.314,
      "step": 30000
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0003492462311557789,
      "loss": 0.0121,
      "step": 30500
    },
    {
      "epoch": 2.07,
      "learning_rate": 0.00034673366834170853,
      "loss": 0.0121,
      "step": 31000
    },
    {
      "epoch": 2.11,
      "learning_rate": 0.0003442211055276382,
      "loss": 0.0122,
      "step": 31500
    },
    {
      "epoch": 2.14,
      "learning_rate": 0.0003417085427135678,
      "loss": 0.012,
      "step": 32000
    },
    {
      "epoch": 2.17,
      "learning_rate": 0.0003391959798994975,
      "loss": 0.0121,
      "step": 32500
    },
    {
      "epoch": 2.17,
      "eval_loss": 0.010623764246702194,
      "eval_runtime": 7.623,
      "eval_samples_per_second": 131.182,
      "eval_steps_per_second": 1.312,
      "step": 32500
    },
    {
      "epoch": 2.21,
      "learning_rate": 0.0003366834170854272,
      "loss": 0.012,
      "step": 33000
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.0003341708542713568,
      "loss": 0.0119,
      "step": 33500
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.00033165829145728645,
      "loss": 0.0119,
      "step": 34000
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.00032914572864321606,
      "loss": 0.0118,
      "step": 34500
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.00032663316582914577,
      "loss": 0.0118,
      "step": 35000
    },
    {
      "epoch": 2.34,
      "eval_loss": 0.01009445358067751,
      "eval_runtime": 7.8168,
      "eval_samples_per_second": 127.93,
      "eval_steps_per_second": 1.279,
      "step": 35000
    },
    {
      "epoch": 2.38,
      "learning_rate": 0.0003241206030150754,
      "loss": 0.0119,
      "step": 35500
    },
    {
      "epoch": 2.41,
      "learning_rate": 0.00032160804020100504,
      "loss": 0.0119,
      "step": 36000
    },
    {
      "epoch": 2.44,
      "learning_rate": 0.00031909547738693465,
      "loss": 0.0117,
      "step": 36500
    },
    {
      "epoch": 2.48,
      "learning_rate": 0.0003165829145728643,
      "loss": 0.0117,
      "step": 37000
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.00031407035175879403,
      "loss": 0.0115,
      "step": 37500
    },
    {
      "epoch": 2.51,
      "eval_loss": 0.010385086759924889,
      "eval_runtime": 8.0071,
      "eval_samples_per_second": 124.889,
      "eval_steps_per_second": 1.249,
      "step": 37500
    },
    {
      "epoch": 2.54,
      "learning_rate": 0.00031155778894472364,
      "loss": 0.0116,
      "step": 38000
    },
    {
      "epoch": 2.58,
      "learning_rate": 0.0003090452261306533,
      "loss": 0.0115,
      "step": 38500
    },
    {
      "epoch": 2.61,
      "learning_rate": 0.0003065326633165829,
      "loss": 0.0115,
      "step": 39000
    },
    {
      "epoch": 2.64,
      "learning_rate": 0.00030402010050251257,
      "loss": 0.0117,
      "step": 39500
    },
    {
      "epoch": 2.68,
      "learning_rate": 0.0003015075376884422,
      "loss": 0.0116,
      "step": 40000
    },
    {
      "epoch": 2.68,
      "eval_loss": 0.010146565735340118,
      "eval_runtime": 7.8031,
      "eval_samples_per_second": 128.154,
      "eval_steps_per_second": 1.282,
      "step": 40000
    },
    {
      "epoch": 2.71,
      "learning_rate": 0.0002989949748743719,
      "loss": 0.0115,
      "step": 40500
    },
    {
      "epoch": 2.74,
      "learning_rate": 0.0002964824120603015,
      "loss": 0.0113,
      "step": 41000
    },
    {
      "epoch": 2.78,
      "learning_rate": 0.00029396984924623116,
      "loss": 0.0112,
      "step": 41500
    },
    {
      "epoch": 2.81,
      "learning_rate": 0.00029145728643216077,
      "loss": 0.0113,
      "step": 42000
    },
    {
      "epoch": 2.84,
      "learning_rate": 0.00028894472361809043,
      "loss": 0.0113,
      "step": 42500
    },
    {
      "epoch": 2.84,
      "eval_loss": 0.009525712579488754,
      "eval_runtime": 7.6057,
      "eval_samples_per_second": 131.48,
      "eval_steps_per_second": 1.315,
      "step": 42500
    },
    {
      "epoch": 2.88,
      "learning_rate": 0.00028643216080402015,
      "loss": 0.0111,
      "step": 43000
    },
    {
      "epoch": 2.91,
      "learning_rate": 0.00028391959798994976,
      "loss": 0.0111,
      "step": 43500
    },
    {
      "epoch": 2.94,
      "learning_rate": 0.0002814070351758794,
      "loss": 0.011,
      "step": 44000
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.000278894472361809,
      "loss": 0.0109,
      "step": 44500
    },
    {
      "epoch": 3.01,
      "learning_rate": 0.0002763819095477387,
      "loss": 0.0106,
      "step": 45000
    },
    {
      "epoch": 3.01,
      "eval_loss": 0.009837072342634201,
      "eval_runtime": 7.6063,
      "eval_samples_per_second": 131.469,
      "eval_steps_per_second": 1.315,
      "step": 45000
    },
    {
      "epoch": 3.04,
      "learning_rate": 0.00027386934673366835,
      "loss": 0.0091,
      "step": 45500
    },
    {
      "epoch": 3.08,
      "learning_rate": 0.000271356783919598,
      "loss": 0.009,
      "step": 46000
    },
    {
      "epoch": 3.11,
      "learning_rate": 0.0002688442211055276,
      "loss": 0.0088,
      "step": 46500
    },
    {
      "epoch": 3.14,
      "learning_rate": 0.0002663316582914573,
      "loss": 0.0091,
      "step": 47000
    },
    {
      "epoch": 3.18,
      "learning_rate": 0.0002638190954773869,
      "loss": 0.0092,
      "step": 47500
    },
    {
      "epoch": 3.18,
      "eval_loss": 0.009356345981359482,
      "eval_runtime": 7.6201,
      "eval_samples_per_second": 131.232,
      "eval_steps_per_second": 1.312,
      "step": 47500
    },
    {
      "epoch": 3.21,
      "learning_rate": 0.0002613065326633166,
      "loss": 0.0092,
      "step": 48000
    },
    {
      "epoch": 3.25,
      "learning_rate": 0.00025879396984924627,
      "loss": 0.009,
      "step": 48500
    },
    {
      "epoch": 3.28,
      "learning_rate": 0.0002562814070351759,
      "loss": 0.0089,
      "step": 49000
    },
    {
      "epoch": 3.31,
      "learning_rate": 0.00025376884422110554,
      "loss": 0.0092,
      "step": 49500
    },
    {
      "epoch": 3.35,
      "learning_rate": 0.00025125628140703515,
      "loss": 0.0091,
      "step": 50000
    },
    {
      "epoch": 3.35,
      "eval_loss": 0.009372641332447529,
      "eval_runtime": 7.5969,
      "eval_samples_per_second": 131.633,
      "eval_steps_per_second": 1.316,
      "step": 50000
    },
    {
      "epoch": 3.38,
      "learning_rate": 0.0002487437185929648,
      "loss": 0.0091,
      "step": 50500
    },
    {
      "epoch": 3.41,
      "learning_rate": 0.00024623115577889447,
      "loss": 0.009,
      "step": 51000
    },
    {
      "epoch": 3.45,
      "learning_rate": 0.00024371859296482413,
      "loss": 0.0093,
      "step": 51500
    },
    {
      "epoch": 3.48,
      "learning_rate": 0.00024120603015075377,
      "loss": 0.0092,
      "step": 52000
    },
    {
      "epoch": 3.51,
      "learning_rate": 0.00023869346733668343,
      "loss": 0.009,
      "step": 52500
    },
    {
      "epoch": 3.51,
      "eval_loss": 0.009138910099864006,
      "eval_runtime": 7.5893,
      "eval_samples_per_second": 131.765,
      "eval_steps_per_second": 1.318,
      "step": 52500
    },
    {
      "epoch": 3.55,
      "learning_rate": 0.00023618090452261307,
      "loss": 0.0091,
      "step": 53000
    },
    {
      "epoch": 3.58,
      "learning_rate": 0.00023366834170854273,
      "loss": 0.0089,
      "step": 53500
    },
    {
      "epoch": 3.61,
      "learning_rate": 0.00023115577889447236,
      "loss": 0.009,
      "step": 54000
    },
    {
      "epoch": 3.65,
      "learning_rate": 0.000228643216080402,
      "loss": 0.0092,
      "step": 54500
    },
    {
      "epoch": 3.68,
      "learning_rate": 0.00022613065326633166,
      "loss": 0.0091,
      "step": 55000
    },
    {
      "epoch": 3.68,
      "eval_loss": 0.00910147000104189,
      "eval_runtime": 7.6234,
      "eval_samples_per_second": 131.175,
      "eval_steps_per_second": 1.312,
      "step": 55000
    },
    {
      "epoch": 3.71,
      "learning_rate": 0.0002236180904522613,
      "loss": 0.0089,
      "step": 55500
    },
    {
      "epoch": 3.75,
      "learning_rate": 0.00022110552763819098,
      "loss": 0.0089,
      "step": 56000
    },
    {
      "epoch": 3.78,
      "learning_rate": 0.00021859296482412062,
      "loss": 0.009,
      "step": 56500
    },
    {
      "epoch": 3.81,
      "learning_rate": 0.00021608040201005025,
      "loss": 0.0091,
      "step": 57000
    },
    {
      "epoch": 3.85,
      "learning_rate": 0.00021356783919597992,
      "loss": 0.0089,
      "step": 57500
    },
    {
      "epoch": 3.85,
      "eval_loss": 0.009140917100012302,
      "eval_runtime": 7.5801,
      "eval_samples_per_second": 131.925,
      "eval_steps_per_second": 1.319,
      "step": 57500
    },
    {
      "epoch": 3.88,
      "learning_rate": 0.00021105527638190955,
      "loss": 0.0089,
      "step": 58000
    },
    {
      "epoch": 3.91,
      "learning_rate": 0.00020854271356783919,
      "loss": 0.0088,
      "step": 58500
    },
    {
      "epoch": 3.95,
      "learning_rate": 0.00020603015075376885,
      "loss": 0.0089,
      "step": 59000
    },
    {
      "epoch": 3.98,
      "learning_rate": 0.00020351758793969848,
      "loss": 0.0088,
      "step": 59500
    },
    {
      "epoch": 4.01,
      "learning_rate": 0.00020100502512562814,
      "loss": 0.0081,
      "step": 60000
    },
    {
      "epoch": 4.01,
      "eval_loss": 0.00933513417840004,
      "eval_runtime": 7.577,
      "eval_samples_per_second": 131.978,
      "eval_steps_per_second": 1.32,
      "step": 60000
    },
    {
      "epoch": 4.05,
      "learning_rate": 0.00019849246231155778,
      "loss": 0.0068,
      "step": 60500
    },
    {
      "epoch": 4.08,
      "learning_rate": 0.00019597989949748744,
      "loss": 0.0068,
      "step": 61000
    },
    {
      "epoch": 4.12,
      "learning_rate": 0.0001934673366834171,
      "loss": 0.007,
      "step": 61500
    },
    {
      "epoch": 4.15,
      "learning_rate": 0.00019095477386934674,
      "loss": 0.0068,
      "step": 62000
    },
    {
      "epoch": 4.18,
      "learning_rate": 0.0001884422110552764,
      "loss": 0.0069,
      "step": 62500
    },
    {
      "epoch": 4.18,
      "eval_loss": 0.009106965735554695,
      "eval_runtime": 7.6103,
      "eval_samples_per_second": 131.401,
      "eval_steps_per_second": 1.314,
      "step": 62500
    },
    {
      "epoch": 4.22,
      "learning_rate": 0.00018592964824120604,
      "loss": 0.0068,
      "step": 63000
    },
    {
      "epoch": 4.25,
      "learning_rate": 0.00018341708542713567,
      "loss": 0.007,
      "step": 63500
    },
    {
      "epoch": 4.28,
      "learning_rate": 0.00018090452261306533,
      "loss": 0.007,
      "step": 64000
    },
    {
      "epoch": 4.32,
      "learning_rate": 0.00017839195979899497,
      "loss": 0.0069,
      "step": 64500
    },
    {
      "epoch": 4.35,
      "learning_rate": 0.00017587939698492463,
      "loss": 0.0069,
      "step": 65000
    },
    {
      "epoch": 4.35,
      "eval_loss": 0.009035669267177582,
      "eval_runtime": 7.7828,
      "eval_samples_per_second": 128.488,
      "eval_steps_per_second": 1.285,
      "step": 65000
    },
    {
      "epoch": 4.38,
      "learning_rate": 0.00017336683417085427,
      "loss": 0.007,
      "step": 65500
    },
    {
      "epoch": 4.42,
      "learning_rate": 0.0001708542713567839,
      "loss": 0.0069,
      "step": 66000
    },
    {
      "epoch": 4.45,
      "learning_rate": 0.0001683417085427136,
      "loss": 0.0069,
      "step": 66500
    },
    {
      "epoch": 4.48,
      "learning_rate": 0.00016582914572864322,
      "loss": 0.0068,
      "step": 67000
    },
    {
      "epoch": 4.52,
      "learning_rate": 0.00016331658291457289,
      "loss": 0.0071,
      "step": 67500
    },
    {
      "epoch": 4.52,
      "eval_loss": 0.009078802540898323,
      "eval_runtime": 7.5849,
      "eval_samples_per_second": 131.841,
      "eval_steps_per_second": 1.318,
      "step": 67500
    },
    {
      "epoch": 4.55,
      "learning_rate": 0.00016080402010050252,
      "loss": 0.007,
      "step": 68000
    },
    {
      "epoch": 4.58,
      "learning_rate": 0.00015829145728643216,
      "loss": 0.0071,
      "step": 68500
    },
    {
      "epoch": 4.62,
      "learning_rate": 0.00015577889447236182,
      "loss": 0.0071,
      "step": 69000
    },
    {
      "epoch": 4.65,
      "learning_rate": 0.00015326633165829145,
      "loss": 0.0072,
      "step": 69500
    },
    {
      "epoch": 4.68,
      "learning_rate": 0.0001507537688442211,
      "loss": 0.0069,
      "step": 70000
    },
    {
      "epoch": 4.68,
      "eval_loss": 0.008746389299631119,
      "eval_runtime": 7.5714,
      "eval_samples_per_second": 132.075,
      "eval_steps_per_second": 1.321,
      "step": 70000
    },
    {
      "epoch": 4.72,
      "learning_rate": 0.00014824120603015075,
      "loss": 0.0071,
      "step": 70500
    },
    {
      "epoch": 4.75,
      "learning_rate": 0.00014572864321608039,
      "loss": 0.007,
      "step": 71000
    },
    {
      "epoch": 4.78,
      "learning_rate": 0.00014321608040201007,
      "loss": 0.007,
      "step": 71500
    },
    {
      "epoch": 4.82,
      "learning_rate": 0.0001407035175879397,
      "loss": 0.007,
      "step": 72000
    },
    {
      "epoch": 4.85,
      "learning_rate": 0.00013819095477386934,
      "loss": 0.0069,
      "step": 72500
    },
    {
      "epoch": 4.85,
      "eval_loss": 0.008641643449664116,
      "eval_runtime": 7.59,
      "eval_samples_per_second": 131.752,
      "eval_steps_per_second": 1.318,
      "step": 72500
    },
    {
      "epoch": 4.88,
      "learning_rate": 0.000135678391959799,
      "loss": 0.007,
      "step": 73000
    },
    {
      "epoch": 4.92,
      "learning_rate": 0.00013316582914572864,
      "loss": 0.007,
      "step": 73500
    },
    {
      "epoch": 4.95,
      "learning_rate": 0.0001306532663316583,
      "loss": 0.0069,
      "step": 74000
    },
    {
      "epoch": 4.99,
      "learning_rate": 0.00012814070351758794,
      "loss": 0.0071,
      "step": 74500
    },
    {
      "epoch": 5.02,
      "learning_rate": 0.00012562814070351757,
      "loss": 0.0058,
      "step": 75000
    },
    {
      "epoch": 5.02,
      "eval_loss": 0.00928833894431591,
      "eval_runtime": 7.8485,
      "eval_samples_per_second": 127.413,
      "eval_steps_per_second": 1.274,
      "step": 75000
    },
    {
      "epoch": 5.05,
      "learning_rate": 0.00012311557788944724,
      "loss": 0.005,
      "step": 75500
    },
    {
      "epoch": 5.09,
      "learning_rate": 0.00012060301507537688,
      "loss": 0.0051,
      "step": 76000
    },
    {
      "epoch": 5.12,
      "learning_rate": 0.00011809045226130653,
      "loss": 0.0053,
      "step": 76500
    },
    {
      "epoch": 5.15,
      "learning_rate": 0.00011557788944723618,
      "loss": 0.0052,
      "step": 77000
    },
    {
      "epoch": 5.19,
      "learning_rate": 0.00011306532663316583,
      "loss": 0.0053,
      "step": 77500
    },
    {
      "epoch": 5.19,
      "eval_loss": 0.009319030679762363,
      "eval_runtime": 7.5874,
      "eval_samples_per_second": 131.797,
      "eval_steps_per_second": 1.318,
      "step": 77500
    },
    {
      "epoch": 5.22,
      "learning_rate": 0.00011055276381909549,
      "loss": 0.0053,
      "step": 78000
    },
    {
      "epoch": 5.25,
      "learning_rate": 0.00010804020100502513,
      "loss": 0.0054,
      "step": 78500
    },
    {
      "epoch": 5.29,
      "learning_rate": 0.00010552763819095478,
      "loss": 0.0052,
      "step": 79000
    },
    {
      "epoch": 5.32,
      "learning_rate": 0.00010301507537688442,
      "loss": 0.0053,
      "step": 79500
    },
    {
      "epoch": 5.35,
      "learning_rate": 0.00010050251256281407,
      "loss": 0.0052,
      "step": 80000
    },
    {
      "epoch": 5.35,
      "eval_loss": 0.008985009975731373,
      "eval_runtime": 7.8274,
      "eval_samples_per_second": 127.757,
      "eval_steps_per_second": 1.278,
      "step": 80000
    },
    {
      "epoch": 5.39,
      "learning_rate": 9.798994974874372e-05,
      "loss": 0.0053,
      "step": 80500
    },
    {
      "epoch": 5.42,
      "learning_rate": 9.547738693467337e-05,
      "loss": 0.0053,
      "step": 81000
    },
    {
      "epoch": 5.45,
      "learning_rate": 9.296482412060302e-05,
      "loss": 0.0052,
      "step": 81500
    },
    {
      "epoch": 5.49,
      "learning_rate": 9.045226130653267e-05,
      "loss": 0.0053,
      "step": 82000
    },
    {
      "epoch": 5.52,
      "learning_rate": 8.793969849246232e-05,
      "loss": 0.0052,
      "step": 82500
    },
    {
      "epoch": 5.52,
      "eval_loss": 0.008979461155831814,
      "eval_runtime": 7.6451,
      "eval_samples_per_second": 130.803,
      "eval_steps_per_second": 1.308,
      "step": 82500
    },
    {
      "epoch": 5.55,
      "learning_rate": 8.542713567839195e-05,
      "loss": 0.0051,
      "step": 83000
    },
    {
      "epoch": 5.59,
      "learning_rate": 8.291457286432161e-05,
      "loss": 0.0051,
      "step": 83500
    },
    {
      "epoch": 5.62,
      "learning_rate": 8.040201005025126e-05,
      "loss": 0.0053,
      "step": 84000
    },
    {
      "epoch": 5.65,
      "learning_rate": 7.788944723618091e-05,
      "loss": 0.0053,
      "step": 84500
    },
    {
      "epoch": 5.69,
      "learning_rate": 7.537688442211054e-05,
      "loss": 0.0052,
      "step": 85000
    },
    {
      "epoch": 5.69,
      "eval_loss": 0.008833312429487705,
      "eval_runtime": 7.5954,
      "eval_samples_per_second": 131.659,
      "eval_steps_per_second": 1.317,
      "step": 85000
    },
    {
      "epoch": 5.72,
      "learning_rate": 7.286432160804019e-05,
      "loss": 0.0053,
      "step": 85500
    },
    {
      "epoch": 5.75,
      "learning_rate": 7.035175879396985e-05,
      "loss": 0.0052,
      "step": 86000
    },
    {
      "epoch": 5.79,
      "learning_rate": 6.78391959798995e-05,
      "loss": 0.0052,
      "step": 86500
    },
    {
      "epoch": 5.82,
      "learning_rate": 6.532663316582915e-05,
      "loss": 0.0052,
      "step": 87000
    },
    {
      "epoch": 5.85,
      "learning_rate": 6.281407035175879e-05,
      "loss": 0.0052,
      "step": 87500
    },
    {
      "epoch": 5.85,
      "eval_loss": 0.00872196163982153,
      "eval_runtime": 7.5919,
      "eval_samples_per_second": 131.72,
      "eval_steps_per_second": 1.317,
      "step": 87500
    },
    {
      "epoch": 5.89,
      "learning_rate": 6.030150753768844e-05,
      "loss": 0.0051,
      "step": 88000
    },
    {
      "epoch": 5.92,
      "learning_rate": 5.778894472361809e-05,
      "loss": 0.0052,
      "step": 88500
    },
    {
      "epoch": 5.96,
      "learning_rate": 5.5276381909547746e-05,
      "loss": 0.0053,
      "step": 89000
    },
    {
      "epoch": 5.99,
      "learning_rate": 5.276381909547739e-05,
      "loss": 0.0052,
      "step": 89500
    },
    {
      "epoch": 6.02,
      "learning_rate": 5.0251256281407036e-05,
      "loss": 0.0043,
      "step": 90000
    },
    {
      "epoch": 6.02,
      "eval_loss": 0.009319793432950974,
      "eval_runtime": 7.6253,
      "eval_samples_per_second": 131.143,
      "eval_steps_per_second": 1.311,
      "step": 90000
    },
    {
      "epoch": 6.06,
      "learning_rate": 4.7738693467336685e-05,
      "loss": 0.0039,
      "step": 90500
    },
    {
      "epoch": 6.09,
      "learning_rate": 4.522613065326633e-05,
      "loss": 0.0039,
      "step": 91000
    },
    {
      "epoch": 6.12,
      "learning_rate": 4.2713567839195975e-05,
      "loss": 0.004,
      "step": 91500
    },
    {
      "epoch": 6.16,
      "learning_rate": 4.020100502512563e-05,
      "loss": 0.004,
      "step": 92000
    },
    {
      "epoch": 6.19,
      "learning_rate": 3.768844221105527e-05,
      "loss": 0.0038,
      "step": 92500
    },
    {
      "epoch": 6.19,
      "eval_loss": 0.009469099342823029,
      "eval_runtime": 7.9167,
      "eval_samples_per_second": 126.315,
      "eval_steps_per_second": 1.263,
      "step": 92500
    },
    {
      "epoch": 6.22,
      "learning_rate": 3.517587939698493e-05,
      "loss": 0.004,
      "step": 93000
    },
    {
      "epoch": 6.26,
      "learning_rate": 3.2663316582914576e-05,
      "loss": 0.0039,
      "step": 93500
    },
    {
      "epoch": 6.29,
      "learning_rate": 3.015075376884422e-05,
      "loss": 0.0039,
      "step": 94000
    },
    {
      "epoch": 6.32,
      "learning_rate": 2.7638190954773873e-05,
      "loss": 0.004,
      "step": 94500
    },
    {
      "epoch": 6.36,
      "learning_rate": 2.5125628140703518e-05,
      "loss": 0.0039,
      "step": 95000
    },
    {
      "epoch": 6.36,
      "eval_loss": 0.009595912881195545,
      "eval_runtime": 7.607,
      "eval_samples_per_second": 131.458,
      "eval_steps_per_second": 1.315,
      "step": 95000
    },
    {
      "epoch": 6.39,
      "learning_rate": 2.2613065326633167e-05,
      "loss": 0.004,
      "step": 95500
    },
    {
      "epoch": 6.42,
      "learning_rate": 2.0100502512562815e-05,
      "loss": 0.0038,
      "step": 96000
    },
    {
      "epoch": 6.46,
      "learning_rate": 1.7587939698492464e-05,
      "loss": 0.0039,
      "step": 96500
    },
    {
      "epoch": 6.49,
      "learning_rate": 1.507537688442211e-05,
      "loss": 0.004,
      "step": 97000
    },
    {
      "epoch": 6.52,
      "learning_rate": 1.2562814070351759e-05,
      "loss": 0.0039,
      "step": 97500
    },
    {
      "epoch": 6.52,
      "eval_loss": 0.009522764943540096,
      "eval_runtime": 7.7116,
      "eval_samples_per_second": 129.674,
      "eval_steps_per_second": 1.297,
      "step": 97500
    }
  ],
  "logging_steps": 500,
  "max_steps": 100000,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 6.676282499901161e+18,
  "trial_name": null,
  "trial_params": null
}