{
  "best_global_step": 9770,
  "best_metric": 4.354217052459717,
  "best_model_checkpoint": "sindhibert_scratch/checkpoint-9770",
  "epoch": 5.0,
  "eval_steps": 977,
  "global_step": 9770,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0512,
      "grad_norm": 18.65768051147461,
      "learning_rate": 9.900000000000002e-06,
      "loss": 77.516953125,
      "step": 100
    },
    {
      "epoch": 0.1024,
      "grad_norm": 13.30678939819336,
      "learning_rate": 1.9900000000000003e-05,
      "loss": 67.7811669921875,
      "step": 200
    },
    {
      "epoch": 0.1536,
      "grad_norm": 9.597126007080078,
      "learning_rate": 2.9900000000000002e-05,
      "loss": 60.4075341796875,
      "step": 300
    },
    {
      "epoch": 0.2048,
      "grad_norm": 8.602116584777832,
      "learning_rate": 3.99e-05,
      "loss": 57.3657763671875,
      "step": 400
    },
    {
      "epoch": 0.256,
      "grad_norm": 9.814536094665527,
      "learning_rate": 4.99e-05,
      "loss": 56.28333984375,
      "step": 500
    },
    {
      "epoch": 0.3072,
      "grad_norm": 11.049890518188477,
      "learning_rate": 5.99e-05,
      "loss": 55.6001611328125,
      "step": 600
    },
    {
      "epoch": 0.3584,
      "grad_norm": 11.56328010559082,
      "learning_rate": 6.99e-05,
      "loss": 55.2201416015625,
      "step": 700
    },
    {
      "epoch": 0.4096,
      "grad_norm": 13.450942993164062,
      "learning_rate": 7.99e-05,
      "loss": 54.76716796875,
      "step": 800
    },
    {
      "epoch": 0.4608,
      "grad_norm": 11.986808776855469,
      "learning_rate": 8.99e-05,
      "loss": 54.2532373046875,
      "step": 900
    },
    {
      "epoch": 0.500224,
      "eval_loss": 6.677970886230469,
      "eval_runtime": 53.9539,
      "eval_samples_per_second": 370.687,
      "eval_steps_per_second": 11.584,
      "step": 977
    },
    {
      "epoch": 0.512,
      "grad_norm": 11.47065258026123,
      "learning_rate": 9.99e-05,
      "loss": 53.8267431640625,
      "step": 1000
    },
    {
      "epoch": 0.5632,
      "grad_norm": 14.508301734924316,
      "learning_rate": 9.887115165336375e-05,
      "loss": 53.4335546875,
      "step": 1100
    },
    {
      "epoch": 0.6144,
      "grad_norm": 13.515727043151855,
      "learning_rate": 9.773090079817561e-05,
      "loss": 53.126240234375,
      "step": 1200
    },
    {
      "epoch": 0.6656,
      "grad_norm": 13.126218795776367,
      "learning_rate": 9.659064994298746e-05,
      "loss": 52.5950341796875,
      "step": 1300
    },
    {
      "epoch": 0.7168,
      "grad_norm": 21.854541778564453,
      "learning_rate": 9.545039908779932e-05,
      "loss": 52.1828076171875,
      "step": 1400
    },
    {
      "epoch": 0.768,
      "grad_norm": 18.417003631591797,
      "learning_rate": 9.431014823261119e-05,
      "loss": 51.821767578125,
      "step": 1500
    },
    {
      "epoch": 0.8192,
      "grad_norm": 12.346606254577637,
      "learning_rate": 9.316989737742304e-05,
      "loss": 51.36373046875,
      "step": 1600
    },
    {
      "epoch": 0.8704,
      "grad_norm": 22.9099063873291,
      "learning_rate": 9.202964652223489e-05,
      "loss": 50.9531591796875,
      "step": 1700
    },
    {
      "epoch": 0.9216,
      "grad_norm": 18.55419158935547,
      "learning_rate": 9.088939566704675e-05,
      "loss": 50.487490234375,
      "step": 1800
    },
    {
      "epoch": 0.9728,
      "grad_norm": 16.247209548950195,
      "learning_rate": 8.974914481185861e-05,
      "loss": 50.037265625,
      "step": 1900
    },
    {
      "epoch": 1.0,
      "eval_loss": 6.126930236816406,
      "eval_runtime": 54.5716,
      "eval_samples_per_second": 366.491,
      "eval_steps_per_second": 11.453,
      "step": 1954
    },
    {
      "epoch": 1.023552,
      "grad_norm": 15.577176094055176,
      "learning_rate": 8.860889395667046e-05,
      "loss": 49.1747216796875,
      "step": 2000
    },
    {
      "epoch": 1.074752,
      "grad_norm": 14.534530639648438,
      "learning_rate": 8.746864310148233e-05,
      "loss": 48.833271484375,
      "step": 2100
    },
    {
      "epoch": 1.125952,
      "grad_norm": 21.150440216064453,
      "learning_rate": 8.632839224629419e-05,
      "loss": 47.9638330078125,
      "step": 2200
    },
    {
      "epoch": 1.177152,
      "grad_norm": 14.35093879699707,
      "learning_rate": 8.518814139110604e-05,
      "loss": 47.6843505859375,
      "step": 2300
    },
    {
      "epoch": 1.228352,
      "grad_norm": 18.37192726135254,
      "learning_rate": 8.40478905359179e-05,
      "loss": 47.1163623046875,
      "step": 2400
    },
    {
      "epoch": 1.279552,
      "grad_norm": 15.366902351379395,
      "learning_rate": 8.290763968072977e-05,
      "loss": 46.6763720703125,
      "step": 2500
    },
    {
      "epoch": 1.330752,
      "grad_norm": 18.58373260498047,
      "learning_rate": 8.176738882554162e-05,
      "loss": 46.071142578125,
      "step": 2600
    },
    {
      "epoch": 1.381952,
      "grad_norm": 16.35076141357422,
      "learning_rate": 8.062713797035348e-05,
      "loss": 45.5451220703125,
      "step": 2700
    },
    {
      "epoch": 1.433152,
      "grad_norm": 18.8570556640625,
      "learning_rate": 7.948688711516535e-05,
      "loss": 45.1469384765625,
      "step": 2800
    },
    {
      "epoch": 1.484352,
      "grad_norm": 17.72637367248535,
      "learning_rate": 7.83466362599772e-05,
      "loss": 44.6100927734375,
      "step": 2900
    },
    {
      "epoch": 1.500224,
      "eval_loss": 5.475055694580078,
      "eval_runtime": 54.4157,
      "eval_samples_per_second": 367.541,
      "eval_steps_per_second": 11.486,
      "step": 2931
    },
    {
      "epoch": 1.535552,
      "grad_norm": 14.798540115356445,
      "learning_rate": 7.720638540478906e-05,
      "loss": 44.670400390625,
      "step": 3000
    },
    {
      "epoch": 1.5867520000000002,
      "grad_norm": 17.05501365661621,
      "learning_rate": 7.606613454960093e-05,
      "loss": 44.11970703125,
      "step": 3100
    },
    {
      "epoch": 1.6379519999999999,
      "grad_norm": 15.10769271850586,
      "learning_rate": 7.492588369441278e-05,
      "loss": 43.6269775390625,
      "step": 3200
    },
    {
      "epoch": 1.689152,
      "grad_norm": 16.199745178222656,
      "learning_rate": 7.378563283922463e-05,
      "loss": 43.4040771484375,
      "step": 3300
    },
    {
      "epoch": 1.7403520000000001,
      "grad_norm": 19.104358673095703,
      "learning_rate": 7.264538198403649e-05,
      "loss": 42.9161328125,
      "step": 3400
    },
    {
      "epoch": 1.791552,
      "grad_norm": 17.44623374938965,
      "learning_rate": 7.150513112884834e-05,
      "loss": 42.778564453125,
      "step": 3500
    },
    {
      "epoch": 1.842752,
      "grad_norm": 16.97149658203125,
      "learning_rate": 7.03648802736602e-05,
      "loss": 42.4598193359375,
      "step": 3600
    },
    {
      "epoch": 1.893952,
      "grad_norm": 14.990788459777832,
      "learning_rate": 6.922462941847207e-05,
      "loss": 42.2105419921875,
      "step": 3700
    },
    {
      "epoch": 1.945152,
      "grad_norm": 17.827381134033203,
      "learning_rate": 6.808437856328392e-05,
      "loss": 42.019033203125,
      "step": 3800
    },
    {
      "epoch": 1.996352,
      "grad_norm": 20.906902313232422,
      "learning_rate": 6.694412770809578e-05,
      "loss": 41.5656201171875,
      "step": 3900
    },
    {
      "epoch": 2.0,
      "eval_loss": 5.107455253601074,
      "eval_runtime": 54.6023,
      "eval_samples_per_second": 366.285,
      "eval_steps_per_second": 11.446,
      "step": 3908
    },
    {
      "epoch": 2.047104,
      "grad_norm": 16.346454620361328,
      "learning_rate": 6.580387685290765e-05,
      "loss": 40.89203369140625,
      "step": 4000
    },
    {
      "epoch": 2.098304,
      "grad_norm": 16.84693145751953,
      "learning_rate": 6.46636259977195e-05,
      "loss": 41.0606884765625,
      "step": 4100
    },
    {
      "epoch": 2.149504,
      "grad_norm": 18.569360733032227,
      "learning_rate": 6.352337514253136e-05,
      "loss": 40.92314697265625,
      "step": 4200
    },
    {
      "epoch": 2.200704,
      "grad_norm": 15.775079727172852,
      "learning_rate": 6.238312428734322e-05,
      "loss": 40.7518408203125,
      "step": 4300
    },
    {
      "epoch": 2.251904,
      "grad_norm": 18.271591186523438,
      "learning_rate": 6.124287343215507e-05,
      "loss": 40.4889990234375,
      "step": 4400
    },
    {
      "epoch": 2.303104,
      "grad_norm": 20.265701293945312,
      "learning_rate": 6.010262257696694e-05,
      "loss": 40.1488427734375,
      "step": 4500
    },
    {
      "epoch": 2.354304,
      "grad_norm": 19.53594398498535,
      "learning_rate": 5.8962371721778794e-05,
      "loss": 40.10376708984375,
      "step": 4600
    },
    {
      "epoch": 2.405504,
      "grad_norm": 19.707582473754883,
      "learning_rate": 5.782212086659066e-05,
      "loss": 39.953662109375,
      "step": 4700
    },
    {
      "epoch": 2.456704,
      "grad_norm": 15.16901683807373,
      "learning_rate": 5.6681870011402515e-05,
      "loss": 39.77469970703125,
      "step": 4800
    },
    {
      "epoch": 2.5002240000000002,
      "eval_loss": 4.868845462799072,
      "eval_runtime": 54.8341,
      "eval_samples_per_second": 364.737,
      "eval_steps_per_second": 11.398,
      "step": 4885
    },
    {
      "epoch": 2.507904,
      "grad_norm": 17.12852668762207,
      "learning_rate": 5.554161915621437e-05,
      "loss": 39.64916015625,
      "step": 4900
    },
    {
      "epoch": 2.559104,
      "grad_norm": 19.869260787963867,
      "learning_rate": 5.440136830102622e-05,
      "loss": 39.51341552734375,
      "step": 5000
    },
    {
      "epoch": 2.610304,
      "grad_norm": 17.342073440551758,
      "learning_rate": 5.326111744583808e-05,
      "loss": 39.350986328125,
      "step": 5100
    },
    {
      "epoch": 2.661504,
      "grad_norm": 19.635601043701172,
      "learning_rate": 5.212086659064994e-05,
      "loss": 39.2056787109375,
      "step": 5200
    },
    {
      "epoch": 2.712704,
      "grad_norm": 16.2427921295166,
      "learning_rate": 5.09806157354618e-05,
      "loss": 38.98402587890625,
      "step": 5300
    },
    {
      "epoch": 2.763904,
      "grad_norm": 21.025632858276367,
      "learning_rate": 4.984036488027366e-05,
      "loss": 38.84795166015625,
      "step": 5400
    },
    {
      "epoch": 2.815104,
      "grad_norm": 16.225173950195312,
      "learning_rate": 4.870011402508552e-05,
      "loss": 38.63864013671875,
      "step": 5500
    },
    {
      "epoch": 2.866304,
      "grad_norm": 19.175825119018555,
      "learning_rate": 4.755986316989738e-05,
      "loss": 38.5738232421875,
      "step": 5600
    },
    {
      "epoch": 2.917504,
      "grad_norm": 18.190704345703125,
      "learning_rate": 4.6419612314709235e-05,
      "loss": 38.47998046875,
      "step": 5700
    },
    {
      "epoch": 2.968704,
      "grad_norm": 17.764493942260742,
      "learning_rate": 4.52793614595211e-05,
      "loss": 38.36706298828125,
      "step": 5800
    },
    {
      "epoch": 3.0,
      "eval_loss": 4.690184116363525,
      "eval_runtime": 54.4789,
      "eval_samples_per_second": 367.115,
      "eval_steps_per_second": 11.472,
      "step": 5862
    },
    {
      "epoch": 3.019456,
      "grad_norm": 16.051603317260742,
      "learning_rate": 4.4139110604332956e-05,
      "loss": 37.86203369140625,
      "step": 5900
    },
    {
      "epoch": 3.070656,
      "grad_norm": 16.511327743530273,
      "learning_rate": 4.299885974914481e-05,
      "loss": 38.09601318359375,
      "step": 6000
    },
    {
      "epoch": 3.121856,
      "grad_norm": 16.089813232421875,
      "learning_rate": 4.1858608893956676e-05,
      "loss": 37.88030029296875,
      "step": 6100
    },
    {
      "epoch": 3.173056,
      "grad_norm": 18.612686157226562,
      "learning_rate": 4.0718358038768533e-05,
      "loss": 37.66849853515625,
      "step": 6200
    },
    {
      "epoch": 3.224256,
      "grad_norm": 17.79659652709961,
      "learning_rate": 3.957810718358039e-05,
      "loss": 37.43648681640625,
      "step": 6300
    },
    {
      "epoch": 3.275456,
      "grad_norm": 16.967939376831055,
      "learning_rate": 3.843785632839225e-05,
      "loss": 37.43171142578125,
      "step": 6400
    },
    {
      "epoch": 3.326656,
      "grad_norm": 16.4842529296875,
      "learning_rate": 3.7297605473204104e-05,
      "loss": 37.26192138671875,
      "step": 6500
    },
    {
      "epoch": 3.377856,
      "grad_norm": 16.261512756347656,
      "learning_rate": 3.615735461801597e-05,
      "loss": 37.26702392578125,
      "step": 6600
    },
    {
      "epoch": 3.429056,
      "grad_norm": 17.182903289794922,
      "learning_rate": 3.5017103762827825e-05,
      "loss": 37.16359619140625,
      "step": 6700
    },
    {
      "epoch": 3.480256,
      "grad_norm": 17.765832901000977,
      "learning_rate": 3.387685290763968e-05,
      "loss": 37.05822998046875,
      "step": 6800
    },
    {
      "epoch": 3.5002240000000002,
      "eval_loss": 4.552316188812256,
      "eval_runtime": 54.6295,
      "eval_samples_per_second": 366.103,
      "eval_steps_per_second": 11.441,
      "step": 6839
    },
    {
      "epoch": 3.531456,
      "grad_norm": 18.267107009887695,
      "learning_rate": 3.2736602052451546e-05,
      "loss": 36.907744140625,
      "step": 6900
    },
    {
      "epoch": 3.582656,
      "grad_norm": 20.562347412109375,
      "learning_rate": 3.15963511972634e-05,
      "loss": 36.8392919921875,
      "step": 7000
    },
    {
      "epoch": 3.6338559999999998,
      "grad_norm": 16.440811157226562,
      "learning_rate": 3.0456100342075257e-05,
      "loss": 36.639384765625,
      "step": 7100
    },
    {
      "epoch": 3.685056,
      "grad_norm": 17.070350646972656,
      "learning_rate": 2.9315849486887114e-05,
      "loss": 36.63670166015625,
      "step": 7200
    },
    {
      "epoch": 3.736256,
      "grad_norm": 17.15755844116211,
      "learning_rate": 2.8175598631698974e-05,
      "loss": 36.6428759765625,
      "step": 7300
    },
    {
      "epoch": 3.787456,
      "grad_norm": 18.988977432250977,
      "learning_rate": 2.7035347776510834e-05,
      "loss": 36.578349609375,
      "step": 7400
    },
    {
      "epoch": 3.838656,
      "grad_norm": 17.87462615966797,
      "learning_rate": 2.589509692132269e-05,
      "loss": 36.4393994140625,
      "step": 7500
    },
    {
      "epoch": 3.889856,
      "grad_norm": 18.244951248168945,
      "learning_rate": 2.4754846066134552e-05,
      "loss": 36.40805908203125,
      "step": 7600
    },
    {
      "epoch": 3.941056,
      "grad_norm": 16.166940689086914,
      "learning_rate": 2.361459521094641e-05,
      "loss": 36.23387451171875,
      "step": 7700
    },
    {
      "epoch": 3.9922560000000002,
      "grad_norm": 19.10250473022461,
      "learning_rate": 2.2474344355758266e-05,
      "loss": 36.08807373046875,
      "step": 7800
    },
    {
      "epoch": 4.0,
      "eval_loss": 4.433136940002441,
      "eval_runtime": 54.6448,
      "eval_samples_per_second": 366.0,
      "eval_steps_per_second": 11.438,
      "step": 7816
    },
    {
      "epoch": 4.043008,
      "grad_norm": 17.705839157104492,
      "learning_rate": 2.1334093500570126e-05,
      "loss": 35.8281103515625,
      "step": 7900
    },
    {
      "epoch": 4.094208,
      "grad_norm": 16.15859031677246,
      "learning_rate": 2.0193842645381987e-05,
      "loss": 35.8823193359375,
      "step": 8000
    },
    {
      "epoch": 4.145408,
      "grad_norm": 17.96906852722168,
      "learning_rate": 1.9053591790193844e-05,
      "loss": 35.968681640625,
      "step": 8100
    },
    {
      "epoch": 4.196608,
      "grad_norm": 16.74631118774414,
      "learning_rate": 1.79133409350057e-05,
      "loss": 35.894248046875,
      "step": 8200
    },
    {
      "epoch": 4.247808,
      "grad_norm": 16.067609786987305,
      "learning_rate": 1.677309007981756e-05,
      "loss": 35.78274169921875,
      "step": 8300
    },
    {
      "epoch": 4.299008,
      "grad_norm": 17.308250427246094,
      "learning_rate": 1.563283922462942e-05,
      "loss": 35.8246630859375,
      "step": 8400
    },
    {
      "epoch": 4.350208,
      "grad_norm": 17.30617332458496,
      "learning_rate": 1.4492588369441278e-05,
      "loss": 35.7324560546875,
      "step": 8500
    },
    {
      "epoch": 4.401408,
      "grad_norm": 16.38850975036621,
      "learning_rate": 1.3352337514253135e-05,
      "loss": 35.88875244140625,
      "step": 8600
    },
    {
      "epoch": 4.452608,
      "grad_norm": 20.229633331298828,
      "learning_rate": 1.2212086659064994e-05,
      "loss": 35.7011962890625,
      "step": 8700
    },
    {
      "epoch": 4.500224,
      "eval_loss": 4.369416236877441,
      "eval_runtime": 54.3995,
      "eval_samples_per_second": 367.651,
      "eval_steps_per_second": 11.489,
      "step": 8793
    },
    {
      "epoch": 4.503808,
      "grad_norm": 18.37810707092285,
      "learning_rate": 1.1071835803876854e-05,
      "loss": 35.69582763671875,
      "step": 8800
    },
    {
      "epoch": 4.555008,
      "grad_norm": 16.187055587768555,
      "learning_rate": 9.931584948688711e-06,
      "loss": 35.73132080078125,
      "step": 8900
    },
    {
      "epoch": 4.606208,
      "grad_norm": 16.89447021484375,
      "learning_rate": 8.79133409350057e-06,
      "loss": 35.47484375,
      "step": 9000
    },
    {
      "epoch": 4.657408,
      "grad_norm": 17.172351837158203,
      "learning_rate": 7.651083238312429e-06,
      "loss": 35.4910205078125,
      "step": 9100
    },
    {
      "epoch": 4.708608,
      "grad_norm": 18.75235939025879,
      "learning_rate": 6.5108323831242875e-06,
      "loss": 35.401552734375,
      "step": 9200
    },
    {
      "epoch": 4.759808,
      "grad_norm": 17.79449462890625,
      "learning_rate": 5.370581527936146e-06,
      "loss": 35.37335693359375,
      "step": 9300
    },
    {
      "epoch": 4.811008,
      "grad_norm": 16.077747344970703,
      "learning_rate": 4.230330672748005e-06,
      "loss": 35.395380859375,
      "step": 9400
    },
    {
      "epoch": 4.862208,
      "grad_norm": 17.572834014892578,
      "learning_rate": 3.0900798175598636e-06,
      "loss": 35.3202685546875,
      "step": 9500
    },
    {
      "epoch": 4.913408,
      "grad_norm": 17.8610897064209,
      "learning_rate": 1.949828962371722e-06,
      "loss": 35.46013916015625,
      "step": 9600
    },
    {
      "epoch": 4.964608,
      "grad_norm": 18.012996673583984,
      "learning_rate": 8.095781071835805e-07,
      "loss": 35.363046875,
      "step": 9700
    },
    {
      "epoch": 5.0,
      "eval_loss": 4.354217052459717,
      "eval_runtime": 54.5977,
      "eval_samples_per_second": 366.316,
      "eval_steps_per_second": 11.447,
      "step": 9770
    }
  ],
  "logging_steps": 100,
  "max_steps": 9770,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 977,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.5802339072e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}