End of training

README.md

@@ -2,11 +2,23 @@
 library_name: transformers
 tags:
 - generated_from_trainer
+datasets:
+- Hartunka/processed_wikitext-103-raw-v1-rand-100
 metrics:
 - accuracy
 model-index:
 - name: bert_base_rand_100_v2
-  results: []
+  results:
+  - task:
+      name: Masked Language Modeling
+      type: fill-mask
+    dataset:
+      name: Hartunka/processed_wikitext-103-raw-v1-rand-100
+      type: Hartunka/processed_wikitext-103-raw-v1-rand-100
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.15197491100186472
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -14,10 +26,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # bert_base_rand_100_v2
 
-This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
+This model is a fine-tuned version of [](https://huggingface.co/) on the Hartunka/processed_wikitext-103-raw-v1-rand-100 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 14.3447
-- Accuracy: 0.1513
+- Loss: 10.7872
+- Accuracy: 0.1520
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 25.0,
+    "eval_accuracy": 0.15197491100186472,
+    "eval_loss": 10.78721809387207,
+    "eval_runtime": 1.0619,
+    "eval_samples": 479,
+    "eval_samples_per_second": 451.072,
+    "eval_steps_per_second": 4.708,
+    "perplexity": 48398.20928962691,
+    "total_flos": 1.5161929390646784e+18,
+    "train_loss": 8.292202761351412,
+    "train_runtime": 25105.6794,
+    "train_samples": 228639,
+    "train_samples_per_second": 227.677,
+    "train_steps_per_second": 2.372
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 25.0,
+    "eval_accuracy": 0.15197491100186472,
+    "eval_loss": 10.78721809387207,
+    "eval_runtime": 1.0619,
+    "eval_samples": 479,
+    "eval_samples_per_second": 451.072,
+    "eval_steps_per_second": 4.708,
+    "perplexity": 48398.20928962691
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 25.0,
+    "total_flos": 1.5161929390646784e+18,
+    "train_loss": 8.292202761351412,
+    "train_runtime": 25105.6794,
+    "train_samples": 228639,
+    "train_samples_per_second": 227.677,
+    "train_steps_per_second": 2.372
+}

trainer_state.json

@@ -0,0 +1,921 @@
+{
+  "best_global_step": 10000,
+  "best_metric": 10.769588470458984,
+  "best_model_checkpoint": "bert_base_rand_100_v2/checkpoint-10000",
+  "epoch": 25.0,
+  "eval_steps": 10000,
+  "global_step": 59550,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.20990764063811923,
+      "grad_norm": 2.901669502258301,
+      "learning_rate": 5e-06,
+      "loss": 13.8235,
+      "step": 500
+    },
+    {
+      "epoch": 0.41981528127623846,
+      "grad_norm": 2.3694849014282227,
+      "learning_rate": 1e-05,
+      "loss": 12.1689,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6297229219143576,
+      "grad_norm": 2.1171717643737793,
+      "learning_rate": 1.5e-05,
+      "loss": 11.6234,
+      "step": 1500
+    },
+    {
+      "epoch": 0.8396305625524769,
+      "grad_norm": 2.043142318725586,
+      "learning_rate": 2e-05,
+      "loss": 11.4717,
+      "step": 2000
+    },
+    {
+      "epoch": 1.0495382031905962,
+      "grad_norm": 2.388023614883423,
+      "learning_rate": 2.5e-05,
+      "loss": 11.3509,
+      "step": 2500
+    },
+    {
+      "epoch": 1.2594458438287153,
+      "grad_norm": 2.2314789295196533,
+      "learning_rate": 3e-05,
+      "loss": 11.2401,
+      "step": 3000
+    },
+    {
+      "epoch": 1.4693534844668346,
+      "grad_norm": 2.266010046005249,
+      "learning_rate": 3.5e-05,
+      "loss": 11.1514,
+      "step": 3500
+    },
+    {
+      "epoch": 1.6792611251049538,
+      "grad_norm": 2.372283935546875,
+      "learning_rate": 4e-05,
+      "loss": 11.0842,
+      "step": 4000
+    },
+    {
+      "epoch": 1.8891687657430731,
+      "grad_norm": 2.4355216026306152,
+      "learning_rate": 4.5e-05,
+      "loss": 11.0266,
+      "step": 4500
+    },
+    {
+      "epoch": 2.0990764063811924,
+      "grad_norm": 2.190211057662964,
+      "learning_rate": 5e-05,
+      "loss": 10.9776,
+      "step": 5000
+    },
+    {
+      "epoch": 2.3089840470193117,
+      "grad_norm": 2.5813465118408203,
+      "learning_rate": 5.500000000000001e-05,
+      "loss": 10.9433,
+      "step": 5500
+    },
+    {
+      "epoch": 2.5188916876574305,
+      "grad_norm": 2.637795925140381,
+      "learning_rate": 6e-05,
+      "loss": 10.908,
+      "step": 6000
+    },
+    {
+      "epoch": 2.72879932829555,
+      "grad_norm": 2.203212022781372,
+      "learning_rate": 6.500000000000001e-05,
+      "loss": 10.8765,
+      "step": 6500
+    },
+    {
+      "epoch": 2.938706968933669,
+      "grad_norm": 2.396450996398926,
+      "learning_rate": 7e-05,
+      "loss": 10.8495,
+      "step": 7000
+    },
+    {
+      "epoch": 3.1486146095717884,
+      "grad_norm": 2.4682557582855225,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 10.8091,
+      "step": 7500
+    },
+    {
+      "epoch": 3.3585222502099077,
+      "grad_norm": 2.704566717147827,
+      "learning_rate": 8e-05,
+      "loss": 10.7873,
+      "step": 8000
+    },
+    {
+      "epoch": 3.568429890848027,
+      "grad_norm": 2.3623745441436768,
+      "learning_rate": 8.5e-05,
+      "loss": 10.7782,
+      "step": 8500
+    },
+    {
+      "epoch": 3.7783375314861463,
+      "grad_norm": 2.6584396362304688,
+      "learning_rate": 9e-05,
+      "loss": 10.7604,
+      "step": 9000
+    },
+    {
+      "epoch": 3.988245172124265,
+      "grad_norm": 2.7074432373046875,
+      "learning_rate": 9.5e-05,
+      "loss": 10.7486,
+      "step": 9500
+    },
+    {
+      "epoch": 4.198152812762385,
+      "grad_norm": 2.9583351612091064,
+      "learning_rate": 0.0001,
+      "loss": 10.6104,
+      "step": 10000
+    },
+    {
+      "epoch": 4.198152812762385,
+      "eval_accuracy": 0.1495867998553104,
+      "eval_loss": 10.769588470458984,
+      "eval_runtime": 1.0981,
+      "eval_samples_per_second": 436.201,
+      "eval_steps_per_second": 4.553,
+      "step": 10000
+    },
+    {
+      "epoch": 4.408060453400504,
+      "grad_norm": 2.6827378273010254,
+      "learning_rate": 9.899091826437942e-05,
+      "loss": 10.6154,
+      "step": 10500
+    },
+    {
+      "epoch": 4.617968094038623,
+      "grad_norm": 3.2582309246063232,
+      "learning_rate": 9.798183652875883e-05,
+      "loss": 10.6123,
+      "step": 11000
+    },
+    {
+      "epoch": 4.827875734676742,
+      "grad_norm": 3.1174557209014893,
+      "learning_rate": 9.697275479313825e-05,
+      "loss": 10.6174,
+      "step": 11500
+    },
+    {
+      "epoch": 5.037783375314861,
+      "grad_norm": 3.2884767055511475,
+      "learning_rate": 9.596367305751767e-05,
+      "loss": 10.5625,
+      "step": 12000
+    },
+    {
+      "epoch": 5.247691015952981,
+      "grad_norm": 3.5405259132385254,
+      "learning_rate": 9.495459132189707e-05,
+      "loss": 10.3547,
+      "step": 12500
+    },
+    {
+      "epoch": 5.4575986565911,
+      "grad_norm": 3.2048330307006836,
+      "learning_rate": 9.394550958627649e-05,
+      "loss": 10.3815,
+      "step": 13000
+    },
+    {
+      "epoch": 5.667506297229219,
+      "grad_norm": 3.4658734798431396,
+      "learning_rate": 9.293642785065591e-05,
+      "loss": 10.4046,
+      "step": 13500
+    },
+    {
+      "epoch": 5.877413937867338,
+      "grad_norm": 3.5260162353515625,
+      "learning_rate": 9.192734611503532e-05,
+      "loss": 10.4204,
+      "step": 14000
+    },
+    {
+      "epoch": 6.087321578505458,
+      "grad_norm": 3.51088285446167,
+      "learning_rate": 9.091826437941474e-05,
+      "loss": 10.2757,
+      "step": 14500
+    },
+    {
+      "epoch": 6.297229219143577,
+      "grad_norm": 3.715435028076172,
+      "learning_rate": 8.990918264379415e-05,
+      "loss": 10.1092,
+      "step": 15000
+    },
+    {
+      "epoch": 6.507136859781696,
+      "grad_norm": 4.197434425354004,
+      "learning_rate": 8.890010090817356e-05,
+      "loss": 10.152,
+      "step": 15500
+    },
+    {
+      "epoch": 6.717044500419815,
+      "grad_norm": 3.8981223106384277,
+      "learning_rate": 8.789101917255298e-05,
+      "loss": 10.187,
+      "step": 16000
+    },
+    {
+      "epoch": 6.926952141057934,
+      "grad_norm": 3.934309482574463,
+      "learning_rate": 8.68819374369324e-05,
+      "loss": 10.208,
+      "step": 16500
+    },
+    {
+      "epoch": 7.136859781696054,
+      "grad_norm": 4.294552326202393,
+      "learning_rate": 8.58728557013118e-05,
+      "loss": 9.9082,
+      "step": 17000
+    },
+    {
+      "epoch": 7.346767422334173,
+      "grad_norm": 4.9949421882629395,
+      "learning_rate": 8.486377396569122e-05,
+      "loss": 9.7752,
+      "step": 17500
+    },
+    {
+      "epoch": 7.5566750629722925,
+      "grad_norm": 4.881385326385498,
+      "learning_rate": 8.385469223007064e-05,
+      "loss": 9.8214,
+      "step": 18000
+    },
+    {
+      "epoch": 7.766582703610411,
+      "grad_norm": 4.930368423461914,
+      "learning_rate": 8.284561049445006e-05,
+      "loss": 9.8631,
+      "step": 18500
+    },
+    {
+      "epoch": 7.97649034424853,
+      "grad_norm": 5.176417350769043,
+      "learning_rate": 8.183652875882947e-05,
+      "loss": 9.8886,
+      "step": 19000
+    },
+    {
+      "epoch": 8.18639798488665,
+      "grad_norm": 5.562292098999023,
+      "learning_rate": 8.082744702320889e-05,
+      "loss": 9.3491,
+      "step": 19500
+    },
+    {
+      "epoch": 8.39630562552477,
+      "grad_norm": 5.809932231903076,
+      "learning_rate": 7.98183652875883e-05,
+      "loss": 9.3294,
+      "step": 20000
+    },
+    {
+      "epoch": 8.39630562552477,
+      "eval_accuracy": 0.15258565055136228,
+      "eval_loss": 11.347984313964844,
+      "eval_runtime": 1.0526,
+      "eval_samples_per_second": 455.081,
+      "eval_steps_per_second": 4.75,
+      "step": 20000
+    },
+    {
+      "epoch": 8.606213266162888,
+      "grad_norm": 6.111857891082764,
+      "learning_rate": 7.880928355196771e-05,
+      "loss": 9.3507,
+      "step": 20500
+    },
+    {
+      "epoch": 8.816120906801007,
+      "grad_norm": 6.15329647064209,
+      "learning_rate": 7.780020181634713e-05,
+      "loss": 9.4085,
+      "step": 21000
+    },
+    {
+      "epoch": 9.026028547439127,
+      "grad_norm": 7.002889156341553,
+      "learning_rate": 7.679112008072655e-05,
+      "loss": 9.3403,
+      "step": 21500
+    },
+    {
+      "epoch": 9.235936188077247,
+      "grad_norm": 6.72435998916626,
+      "learning_rate": 7.578203834510595e-05,
+      "loss": 8.7883,
+      "step": 22000
+    },
+    {
+      "epoch": 9.445843828715365,
+      "grad_norm": 8.087167739868164,
+      "learning_rate": 7.477295660948537e-05,
+      "loss": 8.8251,
+      "step": 22500
+    },
+    {
+      "epoch": 9.655751469353484,
+      "grad_norm": 8.211538314819336,
+      "learning_rate": 7.376387487386479e-05,
+      "loss": 8.885,
+      "step": 23000
+    },
+    {
+      "epoch": 9.865659109991604,
+      "grad_norm": 7.680613994598389,
+      "learning_rate": 7.27547931382442e-05,
+      "loss": 8.9052,
+      "step": 23500
+    },
+    {
+      "epoch": 10.075566750629722,
+      "grad_norm": 7.66122579574585,
+      "learning_rate": 7.174571140262362e-05,
+      "loss": 8.7111,
+      "step": 24000
+    },
+    {
+      "epoch": 10.285474391267842,
+      "grad_norm": 7.407293796539307,
+      "learning_rate": 7.073662966700304e-05,
+      "loss": 8.3415,
+      "step": 24500
+    },
+    {
+      "epoch": 10.495382031905962,
+      "grad_norm": 8.760315895080566,
+      "learning_rate": 6.972754793138244e-05,
+      "loss": 8.3796,
+      "step": 25000
+    },
+    {
+      "epoch": 10.705289672544081,
+      "grad_norm": 7.895341396331787,
+      "learning_rate": 6.871846619576186e-05,
+      "loss": 8.4468,
+      "step": 25500
+    },
+    {
+      "epoch": 10.9151973131822,
+      "grad_norm": 7.624028205871582,
+      "learning_rate": 6.770938446014128e-05,
+      "loss": 8.4814,
+      "step": 26000
+    },
+    {
+      "epoch": 11.125104953820319,
+      "grad_norm": 6.862207412719727,
+      "learning_rate": 6.670030272452068e-05,
+      "loss": 8.1529,
+      "step": 26500
+    },
+    {
+      "epoch": 11.335012594458439,
+      "grad_norm": 10.084607124328613,
+      "learning_rate": 6.56912209889001e-05,
+      "loss": 7.9701,
+      "step": 27000
+    },
+    {
+      "epoch": 11.544920235096557,
+      "grad_norm": 8.635483741760254,
+      "learning_rate": 6.468213925327952e-05,
+      "loss": 8.0247,
+      "step": 27500
+    },
+    {
+      "epoch": 11.754827875734676,
+      "grad_norm": 8.569872856140137,
+      "learning_rate": 6.367305751765893e-05,
+      "loss": 8.0664,
+      "step": 28000
+    },
+    {
+      "epoch": 11.964735516372796,
+      "grad_norm": 8.926000595092773,
+      "learning_rate": 6.266397578203835e-05,
+      "loss": 8.098,
+      "step": 28500
+    },
+    {
+      "epoch": 12.174643157010916,
+      "grad_norm": 6.693151473999023,
+      "learning_rate": 6.165489404641777e-05,
+      "loss": 7.6725,
+      "step": 29000
+    },
+    {
+      "epoch": 12.384550797649034,
+      "grad_norm": 9.540403366088867,
+      "learning_rate": 6.064581231079718e-05,
+      "loss": 7.6162,
+      "step": 29500
+    },
+    {
+      "epoch": 12.594458438287154,
+      "grad_norm": 8.936315536499023,
+      "learning_rate": 5.963673057517659e-05,
+      "loss": 7.6841,
+      "step": 30000
+    },
+    {
+      "epoch": 12.594458438287154,
+      "eval_accuracy": 0.1534544029996922,
+      "eval_loss": 12.822964668273926,
+      "eval_runtime": 1.0475,
+      "eval_samples_per_second": 457.294,
+      "eval_steps_per_second": 4.773,
+      "step": 30000
+    },
+    {
+      "epoch": 12.804366078925273,
+      "grad_norm": 9.013081550598145,
+      "learning_rate": 5.8627648839556e-05,
+      "loss": 7.7262,
+      "step": 30500
+    },
+    {
+      "epoch": 13.014273719563391,
+      "grad_norm": 8.852983474731445,
+      "learning_rate": 5.7618567103935416e-05,
+      "loss": 7.7119,
+      "step": 31000
+    },
+    {
+      "epoch": 13.224181360201511,
+      "grad_norm": 9.840579986572266,
+      "learning_rate": 5.660948536831484e-05,
+      "loss": 7.2769,
+      "step": 31500
+    },
+    {
+      "epoch": 13.43408900083963,
+      "grad_norm": 8.996877670288086,
+      "learning_rate": 5.5600403632694254e-05,
+      "loss": 7.3323,
+      "step": 32000
+    },
+    {
+      "epoch": 13.64399664147775,
+      "grad_norm": 9.106225967407227,
+      "learning_rate": 5.4591321897073666e-05,
+      "loss": 7.3993,
+      "step": 32500
+    },
+    {
+      "epoch": 13.853904282115868,
+      "grad_norm": 11.031023979187012,
+      "learning_rate": 5.358224016145308e-05,
+      "loss": 7.4003,
+      "step": 33000
+    },
+    {
+      "epoch": 14.063811922753988,
+      "grad_norm": 8.555463790893555,
+      "learning_rate": 5.257315842583249e-05,
+      "loss": 7.2998,
+      "step": 33500
+    },
+    {
+      "epoch": 14.273719563392108,
+      "grad_norm": 8.444527626037598,
+      "learning_rate": 5.15640766902119e-05,
+      "loss": 7.0364,
+      "step": 34000
+    },
+    {
+      "epoch": 14.483627204030226,
+      "grad_norm": 8.020197868347168,
+      "learning_rate": 5.055499495459133e-05,
+      "loss": 7.0864,
+      "step": 34500
+    },
+    {
+      "epoch": 14.693534844668346,
+      "grad_norm": 9.18572998046875,
+      "learning_rate": 4.954591321897074e-05,
+      "loss": 7.1191,
+      "step": 35000
+    },
+    {
+      "epoch": 14.903442485306465,
+      "grad_norm": 9.23740005493164,
+      "learning_rate": 4.853683148335015e-05,
+      "loss": 7.1304,
+      "step": 35500
+    },
+    {
+      "epoch": 15.113350125944585,
+      "grad_norm": 10.571009635925293,
+      "learning_rate": 4.7527749747729566e-05,
+      "loss": 6.9682,
+      "step": 36000
+    },
+    {
+      "epoch": 15.323257766582703,
+      "grad_norm": 9.158863067626953,
+      "learning_rate": 4.6518668012108985e-05,
+      "loss": 6.8317,
+      "step": 36500
+    },
+    {
+      "epoch": 15.533165407220823,
+      "grad_norm": 9.79255199432373,
+      "learning_rate": 4.55095862764884e-05,
+      "loss": 6.8596,
+      "step": 37000
+    },
+    {
+      "epoch": 15.743073047858942,
+      "grad_norm": 10.768277168273926,
+      "learning_rate": 4.450050454086781e-05,
+      "loss": 6.9018,
+      "step": 37500
+    },
+    {
+      "epoch": 15.95298068849706,
+      "grad_norm": 12.505277633666992,
+      "learning_rate": 4.349142280524723e-05,
+      "loss": 6.9142,
+      "step": 38000
+    },
+    {
+      "epoch": 16.162888329135182,
+      "grad_norm": 7.873836040496826,
+      "learning_rate": 4.248234106962664e-05,
+      "loss": 6.7081,
+      "step": 38500
+    },
+    {
+      "epoch": 16.3727959697733,
+      "grad_norm": 9.61667537689209,
+      "learning_rate": 4.147325933400605e-05,
+      "loss": 6.6896,
+      "step": 39000
+    },
+    {
+      "epoch": 16.582703610411418,
+      "grad_norm": 10.378979682922363,
+      "learning_rate": 4.046417759838547e-05,
+      "loss": 6.6889,
+      "step": 39500
+    },
+    {
+      "epoch": 16.79261125104954,
+      "grad_norm": 10.515976905822754,
+      "learning_rate": 3.9455095862764884e-05,
+      "loss": 6.7089,
+      "step": 40000
+    },
+    {
+      "epoch": 16.79261125104954,
+      "eval_accuracy": 0.15192889985421107,
+      "eval_loss": 13.659928321838379,
+      "eval_runtime": 1.0378,
+      "eval_samples_per_second": 461.549,
+      "eval_steps_per_second": 4.818,
+      "step": 40000
+    },
+    {
+      "epoch": 17.002518891687657,
+      "grad_norm": 10.031341552734375,
+      "learning_rate": 3.84460141271443e-05,
+      "loss": 6.7274,
+      "step": 40500
+    },
+    {
+      "epoch": 17.212426532325775,
+      "grad_norm": 5.892734050750732,
+      "learning_rate": 3.7436932391523715e-05,
+      "loss": 6.5448,
+      "step": 41000
+    },
+    {
+      "epoch": 17.422334172963897,
+      "grad_norm": 8.581425666809082,
+      "learning_rate": 3.642785065590313e-05,
+      "loss": 6.5561,
+      "step": 41500
+    },
+    {
+      "epoch": 17.632241813602015,
+      "grad_norm": 8.536507606506348,
+      "learning_rate": 3.541876892028255e-05,
+      "loss": 6.573,
+      "step": 42000
+    },
+    {
+      "epoch": 17.842149454240133,
+      "grad_norm": 9.256951332092285,
+      "learning_rate": 3.440968718466196e-05,
+      "loss": 6.5971,
+      "step": 42500
+    },
+    {
+      "epoch": 18.052057094878254,
+      "grad_norm": 10.11147403717041,
+      "learning_rate": 3.340060544904137e-05,
+      "loss": 6.568,
+      "step": 43000
+    },
+    {
+      "epoch": 18.261964735516372,
+      "grad_norm": 6.773597240447998,
+      "learning_rate": 3.239152371342079e-05,
+      "loss": 6.4811,
+      "step": 43500
+    },
+    {
+      "epoch": 18.471872376154494,
+      "grad_norm": 7.478096961975098,
+      "learning_rate": 3.13824419778002e-05,
+      "loss": 6.4904,
+      "step": 44000
+    },
+    {
+      "epoch": 18.68178001679261,
+      "grad_norm": 6.4719767570495605,
+      "learning_rate": 3.0373360242179615e-05,
+      "loss": 6.4999,
+      "step": 44500
+    },
+    {
+      "epoch": 18.89168765743073,
+      "grad_norm": 8.928038597106934,
+      "learning_rate": 2.9364278506559034e-05,
+      "loss": 6.5303,
+      "step": 45000
+    },
+    {
+      "epoch": 19.10159529806885,
+      "grad_norm": 7.1556782722473145,
+      "learning_rate": 2.8355196770938446e-05,
+      "loss": 6.4572,
+      "step": 45500
+    },
+    {
+      "epoch": 19.31150293870697,
+      "grad_norm": 9.890637397766113,
+      "learning_rate": 2.734611503531786e-05,
+      "loss": 6.4306,
+      "step": 46000
+    },
+    {
+      "epoch": 19.521410579345087,
+      "grad_norm": 8.66283130645752,
+      "learning_rate": 2.6337033299697278e-05,
+      "loss": 6.4276,
+      "step": 46500
+    },
+    {
+      "epoch": 19.73131821998321,
+      "grad_norm": 7.5953288078308105,
+      "learning_rate": 2.532795156407669e-05,
+      "loss": 6.4337,
+      "step": 47000
+    },
+    {
+      "epoch": 19.941225860621326,
+      "grad_norm": 7.5974040031433105,
+      "learning_rate": 2.4318869828456106e-05,
+      "loss": 6.4332,
+      "step": 47500
+    },
+    {
+      "epoch": 20.151133501259444,
+      "grad_norm": 7.022908687591553,
+      "learning_rate": 2.330978809283552e-05,
+      "loss": 6.407,
+      "step": 48000
+    },
+    {
+      "epoch": 20.361041141897566,
+      "grad_norm": 10.780464172363281,
+      "learning_rate": 2.2300706357214934e-05,
+      "loss": 6.3859,
+      "step": 48500
+    },
+    {
+      "epoch": 20.570948782535684,
+      "grad_norm": 7.321084022521973,
+      "learning_rate": 2.129162462159435e-05,
+      "loss": 6.377,
+      "step": 49000
+    },
+    {
+      "epoch": 20.7808564231738,
+      "grad_norm": 9.159198760986328,
+      "learning_rate": 2.0282542885973765e-05,
+      "loss": 6.3721,
+      "step": 49500
+    },
+    {
+      "epoch": 20.990764063811923,
+      "grad_norm": 5.452278137207031,
+      "learning_rate": 1.927346115035318e-05,
+      "loss": 6.3859,
+      "step": 50000
+    },
+    {
+      "epoch": 20.990764063811923,
+      "eval_accuracy": 0.1513443799590658,
+      "eval_loss": 14.344673156738281,
+      "eval_runtime": 1.0531,
+      "eval_samples_per_second": 454.861,
+      "eval_steps_per_second": 4.748,
+      "step": 50000
+    },
+    {
+      "epoch": 21.20067170445004,
+      "grad_norm": 8.572796821594238,
+      "learning_rate": 1.8264379414732593e-05,
+      "loss": 6.3359,
+      "step": 50500
+    },
+    {
+      "epoch": 21.410579345088163,
+      "grad_norm": 6.8160576820373535,
+      "learning_rate": 1.725529767911201e-05,
+      "loss": 6.3524,
+      "step": 51000
+    },
+    {
+      "epoch": 21.62048698572628,
+      "grad_norm": 7.5786848068237305,
+      "learning_rate": 1.6246215943491424e-05,
+      "loss": 6.3359,
+      "step": 51500
+    },
+    {
+      "epoch": 21.8303946263644,
+      "grad_norm": 6.643558979034424,
+      "learning_rate": 1.5237134207870837e-05,
+      "loss": 6.3354,
+      "step": 52000
+    },
+    {
+      "epoch": 22.04030226700252,
+      "grad_norm": 5.337532043457031,
+      "learning_rate": 1.4228052472250252e-05,
+      "loss": 6.3373,
+      "step": 52500
+    },
+    {
+      "epoch": 22.250209907640638,
+      "grad_norm": 5.409078121185303,
+      "learning_rate": 1.3218970736629668e-05,
+      "loss": 6.2991,
+      "step": 53000
+    },
+    {
+      "epoch": 22.460117548278756,
+      "grad_norm": 8.76995849609375,
+      "learning_rate": 1.2209889001009082e-05,
+      "loss": 6.3009,
+      "step": 53500
+    },
+    {
+      "epoch": 22.670025188916878,
+      "grad_norm": 8.226763725280762,
+      "learning_rate": 1.1200807265388496e-05,
+      "loss": 6.3081,
+      "step": 54000
+    },
+    {
+      "epoch": 22.879932829554996,
+      "grad_norm": 4.850605487823486,
+      "learning_rate": 1.0191725529767912e-05,
+      "loss": 6.2965,
+      "step": 54500
+    },
+    {
+      "epoch": 23.089840470193113,
+      "grad_norm": 5.159167766571045,
+      "learning_rate": 9.182643794147326e-06,
+      "loss": 6.2943,
+      "step": 55000
+    },
+    {
+      "epoch": 23.299748110831235,
+      "grad_norm": 8.396135330200195,
+      "learning_rate": 8.173562058526741e-06,
+      "loss": 6.2701,
+      "step": 55500
+    },
+    {
+      "epoch": 23.509655751469353,
+      "grad_norm": 6.563684940338135,
+      "learning_rate": 7.164480322906156e-06,
+      "loss": 6.2869,
+      "step": 56000
+    },
+    {
+      "epoch": 23.719563392107474,
+      "grad_norm": 8.230422973632812,
+      "learning_rate": 6.155398587285571e-06,
+      "loss": 6.2586,
+      "step": 56500
+    },
+    {
+      "epoch": 23.929471032745592,
+      "grad_norm": 5.5183634757995605,
+      "learning_rate": 5.146316851664986e-06,
+      "loss": 6.2699,
+      "step": 57000
+    },
+    {
+      "epoch": 24.13937867338371,
+      "grad_norm": 6.817364692687988,
+      "learning_rate": 4.1372351160444e-06,
+      "loss": 6.2553,
+      "step": 57500
+    },
+    {
+      "epoch": 24.349286314021832,
+      "grad_norm": 8.217488288879395,
+      "learning_rate": 3.1281533804238145e-06,
+      "loss": 6.2528,
+      "step": 58000
+    },
+    {
+      "epoch": 24.55919395465995,
+      "grad_norm": 6.756996154785156,
+      "learning_rate": 2.1190716448032293e-06,
+      "loss": 6.2589,
+      "step": 58500
+    },
+    {
+      "epoch": 24.769101595298068,
+      "grad_norm": 5.623405933380127,
+      "learning_rate": 1.109989909182644e-06,
+      "loss": 6.2476,
+      "step": 59000
+    },
+    {
+      "epoch": 24.97900923593619,
+      "grad_norm": 5.834296703338623,
+      "learning_rate": 1.0090817356205853e-07,
+      "loss": 6.2385,
+      "step": 59500
+    },
+    {
+      "epoch": 25.0,
+      "step": 59550,
+      "total_flos": 1.5161929390646784e+18,
+      "train_loss": 8.292202761351412,
+      "train_runtime": 25105.6794,
+      "train_samples_per_second": 227.677,
+      "train_steps_per_second": 2.372
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 59550,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 25,
+  "save_steps": 10000,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.5161929390646784e+18,
+  "train_batch_size": 96,
+  "trial_name": null,
+  "trial_params": null
+}