End of training

README.md

@@ -2,11 +2,23 @@
 library_name: transformers
 tags:
 - generated_from_trainer
+datasets:
+- Hartunka/processed_wikitext-103-raw-v1-km-10
 metrics:
 - accuracy
 model-index:
 - name: bert_base_km_10_v2
-  results: []
+  results:
+  - task:
+      name: Masked Language Modeling
+      type: fill-mask
+    dataset:
+      name: Hartunka/processed_wikitext-103-raw-v1-km-10
+      type: Hartunka/processed_wikitext-103-raw-v1-km-10
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.15561959654178675
 ---
 
 <!-- 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_km_10_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-km-10 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 6.3994
-- Accuracy: 0.1518
+- Loss: 6.3252
+- Accuracy: 0.1556
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 25.0,
+    "eval_accuracy": 0.15561959654178675,
+    "eval_loss": 6.325186252593994,
+    "eval_runtime": 1.0585,
+    "eval_samples": 479,
+    "eval_samples_per_second": 452.547,
+    "eval_steps_per_second": 4.724,
+    "perplexity": 558.461818941483,
+    "total_flos": 1.5149776477860864e+18,
+    "train_loss": 6.212542548319755,
+    "train_runtime": 24740.6956,
+    "train_samples": 228639,
+    "train_samples_per_second": 231.035,
+    "train_steps_per_second": 2.407
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 25.0,
+    "eval_accuracy": 0.15561959654178675,
+    "eval_loss": 6.325186252593994,
+    "eval_runtime": 1.0585,
+    "eval_samples": 479,
+    "eval_samples_per_second": 452.547,
+    "eval_steps_per_second": 4.724,
+    "perplexity": 558.461818941483
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 25.0,
+    "total_flos": 1.5149776477860864e+18,
+    "train_loss": 6.212542548319755,
+    "train_runtime": 24740.6956,
+    "train_samples": 228639,
+    "train_samples_per_second": 231.035,
+    "train_steps_per_second": 2.407
+}

trainer_state.json

@@ -0,0 +1,921 @@
+{
+  "best_global_step": 40000,
+  "best_metric": 6.3062849044799805,
+  "best_model_checkpoint": "bert_base_km_10_v2/checkpoint-40000",
+  "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": 2.9562325477600098,
+      "learning_rate": 5e-06,
+      "loss": 11.0235,
+      "step": 500
+    },
+    {
+      "epoch": 0.41981528127623846,
+      "grad_norm": 17.698453903198242,
+      "learning_rate": 1e-05,
+      "loss": 8.8552,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6297229219143576,
+      "grad_norm": 22.179046630859375,
+      "learning_rate": 1.5e-05,
+      "loss": 8.0013,
+      "step": 1500
+    },
+    {
+      "epoch": 0.8396305625524769,
+      "grad_norm": 13.386920928955078,
+      "learning_rate": 2e-05,
+      "loss": 7.658,
+      "step": 2000
+    },
+    {
+      "epoch": 1.0495382031905962,
+      "grad_norm": 4.086881160736084,
+      "learning_rate": 2.5e-05,
+      "loss": 7.4443,
+      "step": 2500
+    },
+    {
+      "epoch": 1.2594458438287153,
+      "grad_norm": 18.590084075927734,
+      "learning_rate": 3e-05,
+      "loss": 7.2741,
+      "step": 3000
+    },
+    {
+      "epoch": 1.4693534844668346,
+      "grad_norm": 5.107962608337402,
+      "learning_rate": 3.5e-05,
+      "loss": 7.148,
+      "step": 3500
+    },
+    {
+      "epoch": 1.6792611251049538,
+      "grad_norm": 5.024056911468506,
+      "learning_rate": 4e-05,
+      "loss": 7.0377,
+      "step": 4000
+    },
+    {
+      "epoch": 1.8891687657430731,
+      "grad_norm": 2.413734197616577,
+      "learning_rate": 4.5e-05,
+      "loss": 6.9474,
+      "step": 4500
+    },
+    {
+      "epoch": 2.0990764063811924,
+      "grad_norm": 8.736750602722168,
+      "learning_rate": 5e-05,
+      "loss": 6.8616,
+      "step": 5000
+    },
+    {
+      "epoch": 2.3089840470193117,
+      "grad_norm": 2.9648070335388184,
+      "learning_rate": 5.500000000000001e-05,
+      "loss": 6.7957,
+      "step": 5500
+    },
+    {
+      "epoch": 2.5188916876574305,
+      "grad_norm": 2.61800479888916,
+      "learning_rate": 6e-05,
+      "loss": 6.7612,
+      "step": 6000
+    },
+    {
+      "epoch": 2.72879932829555,
+      "grad_norm": 1.6698569059371948,
+      "learning_rate": 6.500000000000001e-05,
+      "loss": 6.7282,
+      "step": 6500
+    },
+    {
+      "epoch": 2.938706968933669,
+      "grad_norm": 3.9698450565338135,
+      "learning_rate": 7e-05,
+      "loss": 6.6944,
+      "step": 7000
+    },
+    {
+      "epoch": 3.1486146095717884,
+      "grad_norm": 3.265909194946289,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 6.6338,
+      "step": 7500
+    },
+    {
+      "epoch": 3.3585222502099077,
+      "grad_norm": 4.147052764892578,
+      "learning_rate": 8e-05,
+      "loss": 6.6197,
+      "step": 8000
+    },
+    {
+      "epoch": 3.568429890848027,
+      "grad_norm": 3.2067461013793945,
+      "learning_rate": 8.5e-05,
+      "loss": 6.6051,
+      "step": 8500
+    },
+    {
+      "epoch": 3.7783375314861463,
+      "grad_norm": 4.531407833099365,
+      "learning_rate": 9e-05,
+      "loss": 6.5858,
+      "step": 9000
+    },
+    {
+      "epoch": 3.988245172124265,
+      "grad_norm": 3.9769198894500732,
+      "learning_rate": 9.5e-05,
+      "loss": 6.5653,
+      "step": 9500
+    },
+    {
+      "epoch": 4.198152812762385,
+      "grad_norm": 2.9494242668151855,
+      "learning_rate": 0.0001,
+      "loss": 6.4969,
+      "step": 10000
+    },
+    {
+      "epoch": 4.198152812762385,
+      "eval_accuracy": 0.1499485238876986,
+      "eval_loss": 6.510183334350586,
+      "eval_runtime": 1.0354,
+      "eval_samples_per_second": 462.609,
+      "eval_steps_per_second": 4.829,
+      "step": 10000
+    },
+    {
+      "epoch": 4.408060453400504,
+      "grad_norm": 1.840803623199463,
+      "learning_rate": 9.899091826437942e-05,
+      "loss": 6.4883,
+      "step": 10500
+    },
+    {
+      "epoch": 4.617968094038623,
+      "grad_norm": 1.7808061838150024,
+      "learning_rate": 9.798183652875883e-05,
+      "loss": 6.4772,
+      "step": 11000
+    },
+    {
+      "epoch": 4.827875734676742,
+      "grad_norm": 2.438516139984131,
+      "learning_rate": 9.697275479313825e-05,
+      "loss": 6.4639,
+      "step": 11500
+    },
+    {
+      "epoch": 5.037783375314861,
+      "grad_norm": 1.6112890243530273,
+      "learning_rate": 9.596367305751767e-05,
+      "loss": 6.4355,
+      "step": 12000
+    },
+    {
+      "epoch": 5.247691015952981,
+      "grad_norm": 3.26106333732605,
+      "learning_rate": 9.495459132189707e-05,
+      "loss": 6.3789,
+      "step": 12500
+    },
+    {
+      "epoch": 5.4575986565911,
+      "grad_norm": 1.8721084594726562,
+      "learning_rate": 9.394550958627649e-05,
+      "loss": 6.3709,
+      "step": 13000
+    },
+    {
+      "epoch": 5.667506297229219,
+      "grad_norm": 1.87173330783844,
+      "learning_rate": 9.293642785065591e-05,
+      "loss": 6.3687,
+      "step": 13500
+    },
+    {
+      "epoch": 5.877413937867338,
+      "grad_norm": 3.479503870010376,
+      "learning_rate": 9.192734611503532e-05,
+      "loss": 6.3636,
+      "step": 14000
+    },
+    {
+      "epoch": 6.087321578505458,
+      "grad_norm": 2.306701421737671,
+      "learning_rate": 9.091826437941474e-05,
+      "loss": 6.323,
+      "step": 14500
+    },
+    {
+      "epoch": 6.297229219143577,
+      "grad_norm": 1.920249581336975,
+      "learning_rate": 8.990918264379415e-05,
+      "loss": 6.2928,
+      "step": 15000
+    },
+    {
+      "epoch": 6.507136859781696,
+      "grad_norm": 1.3836824893951416,
+      "learning_rate": 8.890010090817356e-05,
+      "loss": 6.2833,
+      "step": 15500
+    },
+    {
+      "epoch": 6.717044500419815,
+      "grad_norm": 1.9751460552215576,
+      "learning_rate": 8.789101917255298e-05,
+      "loss": 6.284,
+      "step": 16000
+    },
+    {
+      "epoch": 6.926952141057934,
+      "grad_norm": 1.621890902519226,
+      "learning_rate": 8.68819374369324e-05,
+      "loss": 6.2808,
+      "step": 16500
+    },
+    {
+      "epoch": 7.136859781696054,
+      "grad_norm": 1.5578926801681519,
+      "learning_rate": 8.58728557013118e-05,
+      "loss": 6.2316,
+      "step": 17000
+    },
+    {
+      "epoch": 7.346767422334173,
+      "grad_norm": 1.4018528461456299,
+      "learning_rate": 8.486377396569122e-05,
+      "loss": 6.2145,
+      "step": 17500
+    },
+    {
+      "epoch": 7.5566750629722925,
+      "grad_norm": 2.7806613445281982,
+      "learning_rate": 8.385469223007064e-05,
+      "loss": 6.2183,
+      "step": 18000
+    },
+    {
+      "epoch": 7.766582703610411,
+      "grad_norm": 2.7784783840179443,
+      "learning_rate": 8.284561049445006e-05,
+      "loss": 6.2219,
+      "step": 18500
+    },
+    {
+      "epoch": 7.97649034424853,
+      "grad_norm": 1.9644310474395752,
+      "learning_rate": 8.183652875882947e-05,
+      "loss": 6.2192,
+      "step": 19000
+    },
+    {
+      "epoch": 8.18639798488665,
+      "grad_norm": 2.034841537475586,
+      "learning_rate": 8.082744702320889e-05,
+      "loss": 6.1627,
+      "step": 19500
+    },
+    {
+      "epoch": 8.39630562552477,
+      "grad_norm": 3.0792479515075684,
+      "learning_rate": 7.98183652875883e-05,
+      "loss": 6.1617,
+      "step": 20000
+    },
+    {
+      "epoch": 8.39630562552477,
+      "eval_accuracy": 0.15317489267375628,
+      "eval_loss": 6.349700927734375,
+      "eval_runtime": 1.015,
+      "eval_samples_per_second": 471.903,
+      "eval_steps_per_second": 4.926,
+      "step": 20000
+    },
+    {
+      "epoch": 8.606213266162888,
+      "grad_norm": 1.869396686553955,
+      "learning_rate": 7.880928355196771e-05,
+      "loss": 6.1613,
+      "step": 20500
+    },
+    {
+      "epoch": 8.816120906801007,
+      "grad_norm": 2.249077796936035,
+      "learning_rate": 7.780020181634713e-05,
+      "loss": 6.1601,
+      "step": 21000
+    },
+    {
+      "epoch": 9.026028547439127,
+      "grad_norm": 0.937393844127655,
+      "learning_rate": 7.679112008072655e-05,
+      "loss": 6.1498,
+      "step": 21500
+    },
+    {
+      "epoch": 9.235936188077247,
+      "grad_norm": 1.8309992551803589,
+      "learning_rate": 7.578203834510595e-05,
+      "loss": 6.1159,
+      "step": 22000
+    },
+    {
+      "epoch": 9.445843828715365,
+      "grad_norm": 2.2449676990509033,
+      "learning_rate": 7.477295660948537e-05,
+      "loss": 6.1121,
+      "step": 22500
+    },
+    {
+      "epoch": 9.655751469353484,
+      "grad_norm": 2.006056070327759,
+      "learning_rate": 7.376387487386479e-05,
+      "loss": 6.1132,
+      "step": 23000
+    },
+    {
+      "epoch": 9.865659109991604,
+      "grad_norm": 1.902909517288208,
+      "learning_rate": 7.27547931382442e-05,
+      "loss": 6.1136,
+      "step": 23500
+    },
+    {
+      "epoch": 10.075566750629722,
+      "grad_norm": 1.2427181005477905,
+      "learning_rate": 7.174571140262362e-05,
+      "loss": 6.0892,
+      "step": 24000
+    },
+    {
+      "epoch": 10.285474391267842,
+      "grad_norm": 2.3167381286621094,
+      "learning_rate": 7.073662966700304e-05,
+      "loss": 6.0766,
+      "step": 24500
+    },
+    {
+      "epoch": 10.495382031905962,
+      "grad_norm": 1.573493242263794,
+      "learning_rate": 6.972754793138244e-05,
+      "loss": 6.0668,
+      "step": 25000
+    },
+    {
+      "epoch": 10.705289672544081,
+      "grad_norm": 3.7507879734039307,
+      "learning_rate": 6.871846619576186e-05,
+      "loss": 6.0654,
+      "step": 25500
+    },
+    {
+      "epoch": 10.9151973131822,
+      "grad_norm": 2.302229166030884,
+      "learning_rate": 6.770938446014128e-05,
+      "loss": 6.07,
+      "step": 26000
+    },
+    {
+      "epoch": 11.125104953820319,
+      "grad_norm": 3.5238773822784424,
+      "learning_rate": 6.670030272452068e-05,
+      "loss": 6.0442,
+      "step": 26500
+    },
+    {
+      "epoch": 11.335012594458439,
+      "grad_norm": 2.082115411758423,
+      "learning_rate": 6.56912209889001e-05,
+      "loss": 6.0376,
+      "step": 27000
+    },
+    {
+      "epoch": 11.544920235096557,
+      "grad_norm": 2.6825358867645264,
+      "learning_rate": 6.468213925327952e-05,
+      "loss": 6.0319,
+      "step": 27500
+    },
+    {
+      "epoch": 11.754827875734676,
+      "grad_norm": 2.8914504051208496,
+      "learning_rate": 6.367305751765893e-05,
+      "loss": 6.0314,
+      "step": 28000
+    },
+    {
+      "epoch": 11.964735516372796,
+      "grad_norm": 2.513441801071167,
+      "learning_rate": 6.266397578203835e-05,
+      "loss": 6.0281,
+      "step": 28500
+    },
+    {
+      "epoch": 12.174643157010916,
+      "grad_norm": 2.4295971393585205,
+      "learning_rate": 6.165489404641777e-05,
+      "loss": 6.0116,
+      "step": 29000
+    },
+    {
+      "epoch": 12.384550797649034,
+      "grad_norm": 1.3470953702926636,
+      "learning_rate": 6.064581231079718e-05,
+      "loss": 5.9952,
+      "step": 29500
+    },
+    {
+      "epoch": 12.594458438287154,
+      "grad_norm": 1.2119895219802856,
+      "learning_rate": 5.963673057517659e-05,
+      "loss": 6.0026,
+      "step": 30000
+    },
+    {
+      "epoch": 12.594458438287154,
+      "eval_accuracy": 0.154461762319165,
+      "eval_loss": 6.342120170593262,
+      "eval_runtime": 1.0069,
+      "eval_samples_per_second": 475.715,
+      "eval_steps_per_second": 4.966,
+      "step": 30000
+    },
+    {
+      "epoch": 12.804366078925273,
+      "grad_norm": 2.608718156814575,
+      "learning_rate": 5.8627648839556e-05,
+      "loss": 5.996,
+      "step": 30500
+    },
+    {
+      "epoch": 13.014273719563391,
+      "grad_norm": 1.874100685119629,
+      "learning_rate": 5.7618567103935416e-05,
+      "loss": 5.9987,
+      "step": 31000
+    },
+    {
+      "epoch": 13.224181360201511,
+      "grad_norm": 2.4008915424346924,
+      "learning_rate": 5.660948536831484e-05,
+      "loss": 5.9732,
+      "step": 31500
+    },
+    {
+      "epoch": 13.43408900083963,
+      "grad_norm": 0.8543368577957153,
+      "learning_rate": 5.5600403632694254e-05,
+      "loss": 5.9705,
+      "step": 32000
+    },
+    {
+      "epoch": 13.64399664147775,
+      "grad_norm": 1.8783243894577026,
+      "learning_rate": 5.4591321897073666e-05,
+      "loss": 5.9761,
+      "step": 32500
+    },
+    {
+      "epoch": 13.853904282115868,
+      "grad_norm": 3.0358753204345703,
+      "learning_rate": 5.358224016145308e-05,
+      "loss": 5.9733,
+      "step": 33000
+    },
+    {
+      "epoch": 14.063811922753988,
+      "grad_norm": 2.7263076305389404,
+      "learning_rate": 5.257315842583249e-05,
+      "loss": 5.9591,
+      "step": 33500
+    },
+    {
+      "epoch": 14.273719563392108,
+      "grad_norm": 2.218528985977173,
+      "learning_rate": 5.15640766902119e-05,
+      "loss": 5.9475,
+      "step": 34000
+    },
+    {
+      "epoch": 14.483627204030226,
+      "grad_norm": 2.09617018699646,
+      "learning_rate": 5.055499495459133e-05,
+      "loss": 5.949,
+      "step": 34500
+    },
+    {
+      "epoch": 14.693534844668346,
+      "grad_norm": 0.8091275095939636,
+      "learning_rate": 4.954591321897074e-05,
+      "loss": 5.9477,
+      "step": 35000
+    },
+    {
+      "epoch": 14.903442485306465,
+      "grad_norm": 0.9276658892631531,
+      "learning_rate": 4.853683148335015e-05,
+      "loss": 5.9454,
+      "step": 35500
+    },
+    {
+      "epoch": 15.113350125944585,
+      "grad_norm": 2.462109327316284,
+      "learning_rate": 4.7527749747729566e-05,
+      "loss": 5.9331,
+      "step": 36000
+    },
+    {
+      "epoch": 15.323257766582703,
+      "grad_norm": 2.254075288772583,
+      "learning_rate": 4.6518668012108985e-05,
+      "loss": 5.9298,
+      "step": 36500
+    },
+    {
+      "epoch": 15.533165407220823,
+      "grad_norm": 1.9004478454589844,
+      "learning_rate": 4.55095862764884e-05,
+      "loss": 5.9271,
+      "step": 37000
+    },
+    {
+      "epoch": 15.743073047858942,
+      "grad_norm": 2.358386516571045,
+      "learning_rate": 4.450050454086781e-05,
+      "loss": 5.9278,
+      "step": 37500
+    },
+    {
+      "epoch": 15.95298068849706,
+      "grad_norm": 2.4182186126708984,
+      "learning_rate": 4.349142280524723e-05,
+      "loss": 5.9239,
+      "step": 38000
+    },
+    {
+      "epoch": 16.162888329135182,
+      "grad_norm": 2.934300184249878,
+      "learning_rate": 4.248234106962664e-05,
+      "loss": 5.9158,
+      "step": 38500
+    },
+    {
+      "epoch": 16.3727959697733,
+      "grad_norm": 1.7431899309158325,
+      "learning_rate": 4.147325933400605e-05,
+      "loss": 5.9073,
+      "step": 39000
+    },
+    {
+      "epoch": 16.582703610411418,
+      "grad_norm": 1.3285027742385864,
+      "learning_rate": 4.046417759838547e-05,
+      "loss": 5.908,
+      "step": 39500
+    },
+    {
+      "epoch": 16.79261125104954,
+      "grad_norm": 3.0293195247650146,
+      "learning_rate": 3.9455095862764884e-05,
+      "loss": 5.8999,
+      "step": 40000
+    },
+    {
+      "epoch": 16.79261125104954,
+      "eval_accuracy": 0.15363911629471796,
+      "eval_loss": 6.3062849044799805,
+      "eval_runtime": 1.0052,
+      "eval_samples_per_second": 476.529,
+      "eval_steps_per_second": 4.974,
+      "step": 40000
+    },
+    {
+      "epoch": 17.002518891687657,
+      "grad_norm": 1.3343366384506226,
+      "learning_rate": 3.84460141271443e-05,
+      "loss": 5.9068,
+      "step": 40500
+    },
+    {
+      "epoch": 17.212426532325775,
+      "grad_norm": 0.9144035577774048,
+      "learning_rate": 3.7436932391523715e-05,
+      "loss": 5.8911,
+      "step": 41000
+    },
+    {
+      "epoch": 17.422334172963897,
+      "grad_norm": 1.2867867946624756,
+      "learning_rate": 3.642785065590313e-05,
+      "loss": 5.8886,
+      "step": 41500
+    },
+    {
+      "epoch": 17.632241813602015,
+      "grad_norm": 0.9098787903785706,
+      "learning_rate": 3.541876892028255e-05,
+      "loss": 5.8906,
+      "step": 42000
+    },
+    {
+      "epoch": 17.842149454240133,
+      "grad_norm": 1.5115971565246582,
+      "learning_rate": 3.440968718466196e-05,
+      "loss": 5.8865,
+      "step": 42500
+    },
+    {
+      "epoch": 18.052057094878254,
+      "grad_norm": 0.9073482751846313,
+      "learning_rate": 3.340060544904137e-05,
+      "loss": 5.8811,
+      "step": 43000
+    },
+    {
+      "epoch": 18.261964735516372,
+      "grad_norm": 3.656424045562744,
+      "learning_rate": 3.239152371342079e-05,
+      "loss": 5.8726,
+      "step": 43500
+    },
+    {
+      "epoch": 18.471872376154494,
+      "grad_norm": 2.196565866470337,
+      "learning_rate": 3.13824419778002e-05,
+      "loss": 5.8744,
+      "step": 44000
+    },
+    {
+      "epoch": 18.68178001679261,
+      "grad_norm": 2.9759039878845215,
+      "learning_rate": 3.0373360242179615e-05,
+      "loss": 5.8729,
+      "step": 44500
+    },
+    {
+      "epoch": 18.89168765743073,
+      "grad_norm": 0.7581497430801392,
+      "learning_rate": 2.9364278506559034e-05,
+      "loss": 5.875,
+      "step": 45000
+    },
+    {
+      "epoch": 19.10159529806885,
+      "grad_norm": 2.455521583557129,
+      "learning_rate": 2.8355196770938446e-05,
+      "loss": 5.8716,
+      "step": 45500
+    },
+    {
+      "epoch": 19.31150293870697,
+      "grad_norm": 3.327491283416748,
+      "learning_rate": 2.734611503531786e-05,
+      "loss": 5.858,
+      "step": 46000
+    },
+    {
+      "epoch": 19.521410579345087,
+      "grad_norm": 0.8205088973045349,
+      "learning_rate": 2.6337033299697278e-05,
+      "loss": 5.8592,
+      "step": 46500
+    },
+    {
+      "epoch": 19.73131821998321,
+      "grad_norm": 0.7729224562644958,
+      "learning_rate": 2.532795156407669e-05,
+      "loss": 5.8604,
+      "step": 47000
+    },
+    {
+      "epoch": 19.941225860621326,
+      "grad_norm": 3.09210467338562,
+      "learning_rate": 2.4318869828456106e-05,
+      "loss": 5.8569,
+      "step": 47500
+    },
+    {
+      "epoch": 20.151133501259444,
+      "grad_norm": 0.7637450695037842,
+      "learning_rate": 2.330978809283552e-05,
+      "loss": 5.8508,
+      "step": 48000
+    },
+    {
+      "epoch": 20.361041141897566,
+      "grad_norm": 0.7816163897514343,
+      "learning_rate": 2.2300706357214934e-05,
+      "loss": 5.8493,
+      "step": 48500
+    },
+    {
+      "epoch": 20.570948782535684,
+      "grad_norm": 1.9569042921066284,
+      "learning_rate": 2.129162462159435e-05,
+      "loss": 5.8435,
+      "step": 49000
+    },
+    {
+      "epoch": 20.7808564231738,
+      "grad_norm": 1.819555401802063,
+      "learning_rate": 2.0282542885973765e-05,
+      "loss": 5.8434,
+      "step": 49500
+    },
+    {
+      "epoch": 20.990764063811923,
+      "grad_norm": 3.0075411796569824,
+      "learning_rate": 1.927346115035318e-05,
+      "loss": 5.8424,
+      "step": 50000
+    },
+    {
+      "epoch": 20.990764063811923,
+      "eval_accuracy": 0.1518210110185886,
+      "eval_loss": 6.399431228637695,
+      "eval_runtime": 1.0407,
+      "eval_samples_per_second": 460.263,
+      "eval_steps_per_second": 4.804,
+      "step": 50000
+    },
+    {
+      "epoch": 21.20067170445004,
+      "grad_norm": 1.2910844087600708,
+      "learning_rate": 1.8264379414732593e-05,
+      "loss": 5.8371,
+      "step": 50500
+    },
+    {
+      "epoch": 21.410579345088163,
+      "grad_norm": 1.4958844184875488,
+      "learning_rate": 1.725529767911201e-05,
+      "loss": 5.8351,
+      "step": 51000
+    },
+    {
+      "epoch": 21.62048698572628,
+      "grad_norm": 3.3193728923797607,
+      "learning_rate": 1.6246215943491424e-05,
+      "loss": 5.8341,
+      "step": 51500
+    },
+    {
+      "epoch": 21.8303946263644,
+      "grad_norm": 0.7783209085464478,
+      "learning_rate": 1.5237134207870837e-05,
+      "loss": 5.8346,
+      "step": 52000
+    },
+    {
+      "epoch": 22.04030226700252,
+      "grad_norm": 4.355439186096191,
+      "learning_rate": 1.4228052472250252e-05,
+      "loss": 5.8323,
+      "step": 52500
+    },
+    {
+      "epoch": 22.250209907640638,
+      "grad_norm": 2.6427314281463623,
+      "learning_rate": 1.3218970736629668e-05,
+      "loss": 5.8208,
+      "step": 53000
+    },
+    {
+      "epoch": 22.460117548278756,
+      "grad_norm": 2.9217841625213623,
+      "learning_rate": 1.2209889001009082e-05,
+      "loss": 5.8258,
+      "step": 53500
+    },
+    {
+      "epoch": 22.670025188916878,
+      "grad_norm": 1.1128212213516235,
+      "learning_rate": 1.1200807265388496e-05,
+      "loss": 5.8264,
+      "step": 54000
+    },
+    {
+      "epoch": 22.879932829554996,
+      "grad_norm": 0.7884045243263245,
+      "learning_rate": 1.0191725529767912e-05,
+      "loss": 5.8219,
+      "step": 54500
+    },
+    {
+      "epoch": 23.089840470193113,
+      "grad_norm": 0.7929973602294922,
+      "learning_rate": 9.182643794147326e-06,
+      "loss": 5.8197,
+      "step": 55000
+    },
+    {
+      "epoch": 23.299748110831235,
+      "grad_norm": 0.7606703042984009,
+      "learning_rate": 8.173562058526741e-06,
+      "loss": 5.8172,
+      "step": 55500
+    },
+    {
+      "epoch": 23.509655751469353,
+      "grad_norm": 1.0099503993988037,
+      "learning_rate": 7.164480322906156e-06,
+      "loss": 5.8169,
+      "step": 56000
+    },
+    {
+      "epoch": 23.719563392107474,
+      "grad_norm": 0.7755547165870667,
+      "learning_rate": 6.155398587285571e-06,
+      "loss": 5.8117,
+      "step": 56500
+    },
+    {
+      "epoch": 23.929471032745592,
+      "grad_norm": 0.7879419922828674,
+      "learning_rate": 5.146316851664986e-06,
+      "loss": 5.815,
+      "step": 57000
+    },
+    {
+      "epoch": 24.13937867338371,
+      "grad_norm": 0.7881122827529907,
+      "learning_rate": 4.1372351160444e-06,
+      "loss": 5.8103,
+      "step": 57500
+    },
+    {
+      "epoch": 24.349286314021832,
+      "grad_norm": 0.7837434411048889,
+      "learning_rate": 3.1281533804238145e-06,
+      "loss": 5.8133,
+      "step": 58000
+    },
+    {
+      "epoch": 24.55919395465995,
+      "grad_norm": 0.8001644611358643,
+      "learning_rate": 2.1190716448032293e-06,
+      "loss": 5.8074,
+      "step": 58500
+    },
+    {
+      "epoch": 24.769101595298068,
+      "grad_norm": 0.7615817189216614,
+      "learning_rate": 1.109989909182644e-06,
+      "loss": 5.8077,
+      "step": 59000
+    },
+    {
+      "epoch": 24.97900923593619,
+      "grad_norm": 0.7823144793510437,
+      "learning_rate": 1.0090817356205853e-07,
+      "loss": 5.8079,
+      "step": 59500
+    },
+    {
+      "epoch": 25.0,
+      "step": 59550,
+      "total_flos": 1.5149776477860864e+18,
+      "train_loss": 6.212542548319755,
+      "train_runtime": 24740.6956,
+      "train_samples_per_second": 231.035,
+      "train_steps_per_second": 2.407
+    }
+  ],
+  "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": 1.5149776477860864e+18,
+  "train_batch_size": 96,
+  "trial_name": null,
+  "trial_params": null
+}