{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 22.0,
  "eval_steps": 500,
  "global_step": 10384,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.211864406779661,
      "grad_norm": 2.61049222946167,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 12.9286,
      "step": 100
    },
    {
      "epoch": 0.423728813559322,
      "grad_norm": 2.035546064376831,
      "learning_rate": 4.9518669778296385e-05,
      "loss": 3.184,
      "step": 200
    },
    {
      "epoch": 0.635593220338983,
      "grad_norm": 1.6674128770828247,
      "learning_rate": 4.903247763516142e-05,
      "loss": 2.4283,
      "step": 300
    },
    {
      "epoch": 0.847457627118644,
      "grad_norm": 1.6008472442626953,
      "learning_rate": 4.854628549202645e-05,
      "loss": 2.1681,
      "step": 400
    },
    {
      "epoch": 1.0593220338983051,
      "grad_norm": 1.4100618362426758,
      "learning_rate": 4.806009334889148e-05,
      "loss": 2.0264,
      "step": 500
    },
    {
      "epoch": 1.0593220338983051,
      "eval_loss": 1.4357542991638184,
      "eval_runtime": 1.2436,
      "eval_samples_per_second": 804.114,
      "eval_steps_per_second": 6.433,
      "step": 500
    },
    {
      "epoch": 1.271186440677966,
      "grad_norm": 1.4121861457824707,
      "learning_rate": 4.757390120575652e-05,
      "loss": 1.9105,
      "step": 600
    },
    {
      "epoch": 1.4830508474576272,
      "grad_norm": 1.5169304609298706,
      "learning_rate": 4.708770906262155e-05,
      "loss": 1.8444,
      "step": 700
    },
    {
      "epoch": 1.694915254237288,
      "grad_norm": 1.2196599245071411,
      "learning_rate": 4.660151691948658e-05,
      "loss": 1.7763,
      "step": 800
    },
    {
      "epoch": 1.9067796610169492,
      "grad_norm": 1.2915562391281128,
      "learning_rate": 4.611532477635161e-05,
      "loss": 1.7492,
      "step": 900
    },
    {
      "epoch": 2.1186440677966103,
      "grad_norm": 1.352921485900879,
      "learning_rate": 4.5629132633216645e-05,
      "loss": 1.6981,
      "step": 1000
    },
    {
      "epoch": 2.1186440677966103,
      "eval_loss": 1.2387933731079102,
      "eval_runtime": 1.1755,
      "eval_samples_per_second": 850.716,
      "eval_steps_per_second": 6.806,
      "step": 1000
    },
    {
      "epoch": 2.330508474576271,
      "grad_norm": 1.30990731716156,
      "learning_rate": 4.514294049008168e-05,
      "loss": 1.6452,
      "step": 1100
    },
    {
      "epoch": 2.542372881355932,
      "grad_norm": 1.2262191772460938,
      "learning_rate": 4.4656748346946715e-05,
      "loss": 1.6154,
      "step": 1200
    },
    {
      "epoch": 2.7542372881355934,
      "grad_norm": 1.3794692754745483,
      "learning_rate": 4.4170556203811746e-05,
      "loss": 1.6046,
      "step": 1300
    },
    {
      "epoch": 2.9661016949152543,
      "grad_norm": 1.314780831336975,
      "learning_rate": 4.368436406067678e-05,
      "loss": 1.5812,
      "step": 1400
    },
    {
      "epoch": 3.1779661016949152,
      "grad_norm": 1.1703613996505737,
      "learning_rate": 4.3198171917541816e-05,
      "loss": 1.5312,
      "step": 1500
    },
    {
      "epoch": 3.1779661016949152,
      "eval_loss": 1.1293680667877197,
      "eval_runtime": 1.1994,
      "eval_samples_per_second": 833.764,
      "eval_steps_per_second": 6.67,
      "step": 1500
    },
    {
      "epoch": 3.389830508474576,
      "grad_norm": 1.1206786632537842,
      "learning_rate": 4.271197977440685e-05,
      "loss": 1.5192,
      "step": 1600
    },
    {
      "epoch": 3.601694915254237,
      "grad_norm": 1.0490096807479858,
      "learning_rate": 4.222578763127188e-05,
      "loss": 1.5063,
      "step": 1700
    },
    {
      "epoch": 3.8135593220338984,
      "grad_norm": 1.1094740629196167,
      "learning_rate": 4.173959548813691e-05,
      "loss": 1.4901,
      "step": 1800
    },
    {
      "epoch": 4.02542372881356,
      "grad_norm": 1.0523357391357422,
      "learning_rate": 4.125340334500194e-05,
      "loss": 1.4702,
      "step": 1900
    },
    {
      "epoch": 4.237288135593221,
      "grad_norm": 1.1458876132965088,
      "learning_rate": 4.076721120186698e-05,
      "loss": 1.4313,
      "step": 2000
    },
    {
      "epoch": 4.237288135593221,
      "eval_loss": 1.0662319660186768,
      "eval_runtime": 1.1801,
      "eval_samples_per_second": 847.386,
      "eval_steps_per_second": 6.779,
      "step": 2000
    },
    {
      "epoch": 4.4491525423728815,
      "grad_norm": 1.0316150188446045,
      "learning_rate": 4.028101905873201e-05,
      "loss": 1.438,
      "step": 2100
    },
    {
      "epoch": 4.661016949152542,
      "grad_norm": 1.016991376876831,
      "learning_rate": 3.9794826915597044e-05,
      "loss": 1.4165,
      "step": 2200
    },
    {
      "epoch": 4.872881355932203,
      "grad_norm": 1.2765100002288818,
      "learning_rate": 3.9308634772462075e-05,
      "loss": 1.4133,
      "step": 2300
    },
    {
      "epoch": 5.084745762711864,
      "grad_norm": 1.1391791105270386,
      "learning_rate": 3.882244262932711e-05,
      "loss": 1.3955,
      "step": 2400
    },
    {
      "epoch": 5.296610169491525,
      "grad_norm": 1.0710008144378662,
      "learning_rate": 3.8336250486192145e-05,
      "loss": 1.3629,
      "step": 2500
    },
    {
      "epoch": 5.296610169491525,
      "eval_loss": 1.0101332664489746,
      "eval_runtime": 1.3522,
      "eval_samples_per_second": 739.535,
      "eval_steps_per_second": 5.916,
      "step": 2500
    },
    {
      "epoch": 5.508474576271187,
      "grad_norm": 1.1584999561309814,
      "learning_rate": 3.785005834305718e-05,
      "loss": 1.3656,
      "step": 2600
    },
    {
      "epoch": 5.720338983050848,
      "grad_norm": 0.9541441798210144,
      "learning_rate": 3.736386619992221e-05,
      "loss": 1.3556,
      "step": 2700
    },
    {
      "epoch": 5.932203389830509,
      "grad_norm": 1.0870757102966309,
      "learning_rate": 3.687767405678724e-05,
      "loss": 1.3579,
      "step": 2800
    },
    {
      "epoch": 6.1440677966101696,
      "grad_norm": 1.0170774459838867,
      "learning_rate": 3.639148191365228e-05,
      "loss": 1.3229,
      "step": 2900
    },
    {
      "epoch": 6.3559322033898304,
      "grad_norm": 1.1141579151153564,
      "learning_rate": 3.590528977051731e-05,
      "loss": 1.3049,
      "step": 3000
    },
    {
      "epoch": 6.3559322033898304,
      "eval_loss": 0.9768810272216797,
      "eval_runtime": 1.25,
      "eval_samples_per_second": 799.975,
      "eval_steps_per_second": 6.4,
      "step": 3000
    },
    {
      "epoch": 6.567796610169491,
      "grad_norm": 1.0974624156951904,
      "learning_rate": 3.541909762738234e-05,
      "loss": 1.3239,
      "step": 3100
    },
    {
      "epoch": 6.779661016949152,
      "grad_norm": 1.0595228672027588,
      "learning_rate": 3.493290548424737e-05,
      "loss": 1.3089,
      "step": 3200
    },
    {
      "epoch": 6.991525423728813,
      "grad_norm": 1.0377624034881592,
      "learning_rate": 3.4446713341112405e-05,
      "loss": 1.3133,
      "step": 3300
    },
    {
      "epoch": 7.203389830508475,
      "grad_norm": 1.0550671815872192,
      "learning_rate": 3.396052119797744e-05,
      "loss": 1.2842,
      "step": 3400
    },
    {
      "epoch": 7.415254237288136,
      "grad_norm": 1.097626805305481,
      "learning_rate": 3.3474329054842475e-05,
      "loss": 1.281,
      "step": 3500
    },
    {
      "epoch": 7.415254237288136,
      "eval_loss": 0.9405697584152222,
      "eval_runtime": 1.2881,
      "eval_samples_per_second": 776.366,
      "eval_steps_per_second": 6.211,
      "step": 3500
    },
    {
      "epoch": 7.627118644067797,
      "grad_norm": 0.9320990443229675,
      "learning_rate": 3.2988136911707506e-05,
      "loss": 1.2664,
      "step": 3600
    },
    {
      "epoch": 7.838983050847458,
      "grad_norm": 1.0325995683670044,
      "learning_rate": 3.250194476857254e-05,
      "loss": 1.2701,
      "step": 3700
    },
    {
      "epoch": 8.05084745762712,
      "grad_norm": 0.9588521718978882,
      "learning_rate": 3.2015752625437576e-05,
      "loss": 1.2543,
      "step": 3800
    },
    {
      "epoch": 8.26271186440678,
      "grad_norm": 0.9852187037467957,
      "learning_rate": 3.152956048230261e-05,
      "loss": 1.2419,
      "step": 3900
    },
    {
      "epoch": 8.474576271186441,
      "grad_norm": 1.1718206405639648,
      "learning_rate": 3.104336833916764e-05,
      "loss": 1.2474,
      "step": 4000
    },
    {
      "epoch": 8.474576271186441,
      "eval_loss": 0.9118660092353821,
      "eval_runtime": 1.2362,
      "eval_samples_per_second": 808.929,
      "eval_steps_per_second": 6.471,
      "step": 4000
    },
    {
      "epoch": 8.686440677966102,
      "grad_norm": 1.0222992897033691,
      "learning_rate": 3.055717619603267e-05,
      "loss": 1.2405,
      "step": 4100
    },
    {
      "epoch": 8.898305084745763,
      "grad_norm": 1.1910102367401123,
      "learning_rate": 3.0070984052897706e-05,
      "loss": 1.2345,
      "step": 4200
    },
    {
      "epoch": 9.110169491525424,
      "grad_norm": 0.9338302612304688,
      "learning_rate": 2.9584791909762737e-05,
      "loss": 1.2213,
      "step": 4300
    },
    {
      "epoch": 9.322033898305085,
      "grad_norm": 4.818306922912598,
      "learning_rate": 2.9098599766627772e-05,
      "loss": 1.2286,
      "step": 4400
    },
    {
      "epoch": 9.533898305084746,
      "grad_norm": 1.076406478881836,
      "learning_rate": 2.8612407623492804e-05,
      "loss": 1.2222,
      "step": 4500
    },
    {
      "epoch": 9.533898305084746,
      "eval_loss": 0.8947927951812744,
      "eval_runtime": 1.0672,
      "eval_samples_per_second": 936.991,
      "eval_steps_per_second": 7.496,
      "step": 4500
    },
    {
      "epoch": 9.745762711864407,
      "grad_norm": 0.9330114126205444,
      "learning_rate": 2.8126215480357835e-05,
      "loss": 1.2152,
      "step": 4600
    },
    {
      "epoch": 9.957627118644067,
      "grad_norm": 1.08220374584198,
      "learning_rate": 2.764002333722287e-05,
      "loss": 1.2064,
      "step": 4700
    },
    {
      "epoch": 10.169491525423728,
      "grad_norm": 1.078016757965088,
      "learning_rate": 2.7153831194087902e-05,
      "loss": 1.1838,
      "step": 4800
    },
    {
      "epoch": 10.38135593220339,
      "grad_norm": 1.0999090671539307,
      "learning_rate": 2.6667639050952937e-05,
      "loss": 1.1864,
      "step": 4900
    },
    {
      "epoch": 10.59322033898305,
      "grad_norm": 1.0185160636901855,
      "learning_rate": 2.618144690781797e-05,
      "loss": 1.1951,
      "step": 5000
    },
    {
      "epoch": 10.59322033898305,
      "eval_loss": 0.8726407885551453,
      "eval_runtime": 1.1196,
      "eval_samples_per_second": 893.186,
      "eval_steps_per_second": 7.145,
      "step": 5000
    },
    {
      "epoch": 10.805084745762711,
      "grad_norm": 1.079840064048767,
      "learning_rate": 2.5695254764683003e-05,
      "loss": 1.1893,
      "step": 5100
    },
    {
      "epoch": 11.016949152542374,
      "grad_norm": 1.0007730722427368,
      "learning_rate": 2.5209062621548035e-05,
      "loss": 1.1858,
      "step": 5200
    },
    {
      "epoch": 11.228813559322035,
      "grad_norm": 1.0986075401306152,
      "learning_rate": 2.472287047841307e-05,
      "loss": 1.1656,
      "step": 5300
    },
    {
      "epoch": 11.440677966101696,
      "grad_norm": 0.9882263541221619,
      "learning_rate": 2.4236678335278105e-05,
      "loss": 1.165,
      "step": 5400
    },
    {
      "epoch": 11.652542372881356,
      "grad_norm": 1.046499252319336,
      "learning_rate": 2.3750486192143136e-05,
      "loss": 1.163,
      "step": 5500
    },
    {
      "epoch": 11.652542372881356,
      "eval_loss": 0.8517504930496216,
      "eval_runtime": 1.0586,
      "eval_samples_per_second": 944.64,
      "eval_steps_per_second": 7.557,
      "step": 5500
    },
    {
      "epoch": 11.864406779661017,
      "grad_norm": 0.9540964961051941,
      "learning_rate": 2.326429404900817e-05,
      "loss": 1.1651,
      "step": 5600
    },
    {
      "epoch": 12.076271186440678,
      "grad_norm": 0.9471459984779358,
      "learning_rate": 2.2778101905873203e-05,
      "loss": 1.1531,
      "step": 5700
    },
    {
      "epoch": 12.288135593220339,
      "grad_norm": 1.036387324333191,
      "learning_rate": 2.2291909762738235e-05,
      "loss": 1.1412,
      "step": 5800
    },
    {
      "epoch": 12.5,
      "grad_norm": 0.9427016377449036,
      "learning_rate": 2.180571761960327e-05,
      "loss": 1.1487,
      "step": 5900
    },
    {
      "epoch": 12.711864406779661,
      "grad_norm": 1.1486104726791382,
      "learning_rate": 2.13195254764683e-05,
      "loss": 1.1446,
      "step": 6000
    },
    {
      "epoch": 12.711864406779661,
      "eval_loss": 0.8369362950325012,
      "eval_runtime": 1.6466,
      "eval_samples_per_second": 607.325,
      "eval_steps_per_second": 4.859,
      "step": 6000
    },
    {
      "epoch": 12.923728813559322,
      "grad_norm": 1.1171337366104126,
      "learning_rate": 2.0833333333333336e-05,
      "loss": 1.1481,
      "step": 6100
    },
    {
      "epoch": 13.135593220338983,
      "grad_norm": 0.9343823194503784,
      "learning_rate": 2.0347141190198368e-05,
      "loss": 1.1327,
      "step": 6200
    },
    {
      "epoch": 13.347457627118644,
      "grad_norm": 1.0053530931472778,
      "learning_rate": 1.9860949047063403e-05,
      "loss": 1.1321,
      "step": 6300
    },
    {
      "epoch": 13.559322033898304,
      "grad_norm": 1.0828378200531006,
      "learning_rate": 1.9374756903928434e-05,
      "loss": 1.1286,
      "step": 6400
    },
    {
      "epoch": 13.771186440677965,
      "grad_norm": 1.0227911472320557,
      "learning_rate": 1.8888564760793466e-05,
      "loss": 1.1203,
      "step": 6500
    },
    {
      "epoch": 13.771186440677965,
      "eval_loss": 0.821231484413147,
      "eval_runtime": 1.0368,
      "eval_samples_per_second": 964.483,
      "eval_steps_per_second": 7.716,
      "step": 6500
    },
    {
      "epoch": 13.983050847457626,
      "grad_norm": 1.0566082000732422,
      "learning_rate": 1.84023726176585e-05,
      "loss": 1.1281,
      "step": 6600
    },
    {
      "epoch": 14.194915254237289,
      "grad_norm": 0.9336996674537659,
      "learning_rate": 1.7916180474523532e-05,
      "loss": 1.1172,
      "step": 6700
    },
    {
      "epoch": 14.40677966101695,
      "grad_norm": 0.9151898622512817,
      "learning_rate": 1.7429988331388567e-05,
      "loss": 1.1215,
      "step": 6800
    },
    {
      "epoch": 14.61864406779661,
      "grad_norm": 0.977746844291687,
      "learning_rate": 1.69437961882536e-05,
      "loss": 1.1087,
      "step": 6900
    },
    {
      "epoch": 14.830508474576272,
      "grad_norm": 0.8589827418327332,
      "learning_rate": 1.6457604045118634e-05,
      "loss": 1.1156,
      "step": 7000
    },
    {
      "epoch": 14.830508474576272,
      "eval_loss": 0.8085836172103882,
      "eval_runtime": 1.2389,
      "eval_samples_per_second": 807.162,
      "eval_steps_per_second": 6.457,
      "step": 7000
    },
    {
      "epoch": 15.042372881355933,
      "grad_norm": 1.0282628536224365,
      "learning_rate": 1.5971411901983665e-05,
      "loss": 1.1027,
      "step": 7100
    },
    {
      "epoch": 15.254237288135593,
      "grad_norm": 1.0358235836029053,
      "learning_rate": 1.54852197588487e-05,
      "loss": 1.0936,
      "step": 7200
    },
    {
      "epoch": 15.466101694915254,
      "grad_norm": 0.9948374032974243,
      "learning_rate": 1.4999027615713732e-05,
      "loss": 1.1028,
      "step": 7300
    },
    {
      "epoch": 15.677966101694915,
      "grad_norm": 1.0176063776016235,
      "learning_rate": 1.4512835472578765e-05,
      "loss": 1.1022,
      "step": 7400
    },
    {
      "epoch": 15.889830508474576,
      "grad_norm": 0.930158793926239,
      "learning_rate": 1.4026643329443798e-05,
      "loss": 1.1037,
      "step": 7500
    },
    {
      "epoch": 15.889830508474576,
      "eval_loss": 0.8010894060134888,
      "eval_runtime": 1.0953,
      "eval_samples_per_second": 913.012,
      "eval_steps_per_second": 7.304,
      "step": 7500
    },
    {
      "epoch": 16.10169491525424,
      "grad_norm": 1.0642187595367432,
      "learning_rate": 1.354045118630883e-05,
      "loss": 1.0961,
      "step": 7600
    },
    {
      "epoch": 16.3135593220339,
      "grad_norm": 1.011250615119934,
      "learning_rate": 1.3054259043173863e-05,
      "loss": 1.0911,
      "step": 7700
    },
    {
      "epoch": 16.52542372881356,
      "grad_norm": 0.8920814990997314,
      "learning_rate": 1.2568066900038896e-05,
      "loss": 1.0892,
      "step": 7800
    },
    {
      "epoch": 16.73728813559322,
      "grad_norm": 1.063219666481018,
      "learning_rate": 1.2081874756903928e-05,
      "loss": 1.0847,
      "step": 7900
    },
    {
      "epoch": 16.949152542372882,
      "grad_norm": 0.9891595840454102,
      "learning_rate": 1.1595682613768961e-05,
      "loss": 1.0854,
      "step": 8000
    },
    {
      "epoch": 16.949152542372882,
      "eval_loss": 0.789776086807251,
      "eval_runtime": 1.2524,
      "eval_samples_per_second": 798.447,
      "eval_steps_per_second": 6.388,
      "step": 8000
    },
    {
      "epoch": 17.161016949152543,
      "grad_norm": 1.0095338821411133,
      "learning_rate": 1.1109490470633994e-05,
      "loss": 1.0801,
      "step": 8100
    },
    {
      "epoch": 17.372881355932204,
      "grad_norm": 1.022760272026062,
      "learning_rate": 1.0623298327499028e-05,
      "loss": 1.0792,
      "step": 8200
    },
    {
      "epoch": 17.584745762711865,
      "grad_norm": 1.0345150232315063,
      "learning_rate": 1.0137106184364061e-05,
      "loss": 1.0725,
      "step": 8300
    },
    {
      "epoch": 17.796610169491526,
      "grad_norm": 1.1157786846160889,
      "learning_rate": 9.650914041229094e-06,
      "loss": 1.0827,
      "step": 8400
    },
    {
      "epoch": 18.008474576271187,
      "grad_norm": 1.0153229236602783,
      "learning_rate": 9.164721898094126e-06,
      "loss": 1.0866,
      "step": 8500
    },
    {
      "epoch": 18.008474576271187,
      "eval_loss": 0.7794539928436279,
      "eval_runtime": 1.2458,
      "eval_samples_per_second": 802.72,
      "eval_steps_per_second": 6.422,
      "step": 8500
    },
    {
      "epoch": 18.220338983050848,
      "grad_norm": 1.0266518592834473,
      "learning_rate": 8.678529754959159e-06,
      "loss": 1.0669,
      "step": 8600
    },
    {
      "epoch": 18.43220338983051,
      "grad_norm": 1.039871096611023,
      "learning_rate": 8.192337611824192e-06,
      "loss": 1.0691,
      "step": 8700
    },
    {
      "epoch": 18.64406779661017,
      "grad_norm": 0.9622082114219666,
      "learning_rate": 7.706145468689227e-06,
      "loss": 1.071,
      "step": 8800
    },
    {
      "epoch": 18.85593220338983,
      "grad_norm": 0.9962339997291565,
      "learning_rate": 7.21995332555426e-06,
      "loss": 1.0789,
      "step": 8900
    },
    {
      "epoch": 19.06779661016949,
      "grad_norm": 0.9597769975662231,
      "learning_rate": 6.733761182419293e-06,
      "loss": 1.0712,
      "step": 9000
    },
    {
      "epoch": 19.06779661016949,
      "eval_loss": 0.776411235332489,
      "eval_runtime": 1.274,
      "eval_samples_per_second": 784.919,
      "eval_steps_per_second": 6.279,
      "step": 9000
    },
    {
      "epoch": 19.279661016949152,
      "grad_norm": 1.058910846710205,
      "learning_rate": 6.247569039284325e-06,
      "loss": 1.0732,
      "step": 9100
    },
    {
      "epoch": 19.491525423728813,
      "grad_norm": 1.1111668348312378,
      "learning_rate": 5.761376896149359e-06,
      "loss": 1.0611,
      "step": 9200
    },
    {
      "epoch": 19.703389830508474,
      "grad_norm": 1.0422521829605103,
      "learning_rate": 5.275184753014392e-06,
      "loss": 1.0636,
      "step": 9300
    },
    {
      "epoch": 19.915254237288135,
      "grad_norm": 0.9715420603752136,
      "learning_rate": 4.788992609879424e-06,
      "loss": 1.0623,
      "step": 9400
    },
    {
      "epoch": 20.127118644067796,
      "grad_norm": 1.067021369934082,
      "learning_rate": 4.302800466744458e-06,
      "loss": 1.0648,
      "step": 9500
    },
    {
      "epoch": 20.127118644067796,
      "eval_loss": 0.773140013217926,
      "eval_runtime": 1.0537,
      "eval_samples_per_second": 949.028,
      "eval_steps_per_second": 7.592,
      "step": 9500
    },
    {
      "epoch": 20.338983050847457,
      "grad_norm": 0.9910571575164795,
      "learning_rate": 3.816608323609491e-06,
      "loss": 1.0578,
      "step": 9600
    },
    {
      "epoch": 20.550847457627118,
      "grad_norm": 1.1001310348510742,
      "learning_rate": 3.3304161804745237e-06,
      "loss": 1.0628,
      "step": 9700
    },
    {
      "epoch": 20.76271186440678,
      "grad_norm": 0.9902365803718567,
      "learning_rate": 2.8442240373395565e-06,
      "loss": 1.0598,
      "step": 9800
    },
    {
      "epoch": 20.97457627118644,
      "grad_norm": 1.0591098070144653,
      "learning_rate": 2.3580318942045898e-06,
      "loss": 1.0593,
      "step": 9900
    },
    {
      "epoch": 21.1864406779661,
      "grad_norm": 1.05185067653656,
      "learning_rate": 1.8718397510696226e-06,
      "loss": 1.0551,
      "step": 10000
    },
    {
      "epoch": 21.1864406779661,
      "eval_loss": 0.770883321762085,
      "eval_runtime": 1.265,
      "eval_samples_per_second": 790.545,
      "eval_steps_per_second": 6.324,
      "step": 10000
    },
    {
      "epoch": 21.39830508474576,
      "grad_norm": 0.9792327880859375,
      "learning_rate": 1.3856476079346559e-06,
      "loss": 1.0672,
      "step": 10100
    },
    {
      "epoch": 21.610169491525422,
      "grad_norm": 1.1287437677383423,
      "learning_rate": 8.994554647996889e-07,
      "loss": 1.0577,
      "step": 10200
    },
    {
      "epoch": 21.822033898305087,
      "grad_norm": 1.1712942123413086,
      "learning_rate": 4.13263321664722e-07,
      "loss": 1.0585,
      "step": 10300
    }
  ],
  "logging_steps": 100,
  "max_steps": 10384,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 22,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.582851649954775e+17,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}