End of training

README.md

@@ -2,11 +2,23 @@
 library_name: transformers
 tags:
 - generated_from_trainer
+datasets:
+- Hartunka/processed_wikitext-103-raw-v1-km-20
 metrics:
 - accuracy
 model-index:
 - name: bert_base_km_20_v2
-  results: []
+  results:
+  - task:
+      name: Masked Language Modeling
+      type: fill-mask
+    dataset:
+      name: Hartunka/processed_wikitext-103-raw-v1-km-20
+      type: Hartunka/processed_wikitext-103-raw-v1-km-20
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.15400915409391422
 ---
 
 <!-- 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_20_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-20 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 7.0059
-- Accuracy: 0.1531
+- Loss: 6.7230
+- Accuracy: 0.1540
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 25.0,
+    "eval_accuracy": 0.15400915409391422,
+    "eval_loss": 6.723025321960449,
+    "eval_runtime": 1.0283,
+    "eval_samples": 479,
+    "eval_samples_per_second": 465.802,
+    "eval_steps_per_second": 4.862,
+    "perplexity": 831.3287480343491,
+    "total_flos": 1.5151126801503744e+18,
+    "train_loss": 6.378149575490495,
+    "train_runtime": 24807.4865,
+    "train_samples": 228639,
+    "train_samples_per_second": 230.413,
+    "train_steps_per_second": 2.4
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 25.0,
+    "eval_accuracy": 0.15400915409391422,
+    "eval_loss": 6.723025321960449,
+    "eval_runtime": 1.0283,
+    "eval_samples": 479,
+    "eval_samples_per_second": 465.802,
+    "eval_steps_per_second": 4.862,
+    "perplexity": 831.3287480343491
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 25.0,
+    "total_flos": 1.5151126801503744e+18,
+    "train_loss": 6.378149575490495,
+    "train_runtime": 24807.4865,
+    "train_samples": 228639,
+    "train_samples_per_second": 230.413,
+    "train_steps_per_second": 2.4
+}

trainer_state.json

@@ -0,0 +1,921 @@
+{
+  "best_global_step": 20000,
+  "best_metric": 6.701448440551758,
+  "best_model_checkpoint": "bert_base_km_20_v2/checkpoint-20000",
+  "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": 3.2586686611175537,
+      "learning_rate": 5e-06,
+      "loss": 12.1717,
+      "step": 500
+    },
+    {
+      "epoch": 0.41981528127623846,
+      "grad_norm": 4.922037601470947,
+      "learning_rate": 1e-05,
+      "loss": 10.0311,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6297229219143576,
+      "grad_norm": 28.822181701660156,
+      "learning_rate": 1.5e-05,
+      "loss": 8.5295,
+      "step": 1500
+    },
+    {
+      "epoch": 0.8396305625524769,
+      "grad_norm": 13.167506217956543,
+      "learning_rate": 2e-05,
+      "loss": 8.0722,
+      "step": 2000
+    },
+    {
+      "epoch": 1.0495382031905962,
+      "grad_norm": 16.0736141204834,
+      "learning_rate": 2.5e-05,
+      "loss": 7.8283,
+      "step": 2500
+    },
+    {
+      "epoch": 1.2594458438287153,
+      "grad_norm": 14.928688049316406,
+      "learning_rate": 3e-05,
+      "loss": 7.6626,
+      "step": 3000
+    },
+    {
+      "epoch": 1.4693534844668346,
+      "grad_norm": 6.039346218109131,
+      "learning_rate": 3.5e-05,
+      "loss": 7.5058,
+      "step": 3500
+    },
+    {
+      "epoch": 1.6792611251049538,
+      "grad_norm": 12.346041679382324,
+      "learning_rate": 4e-05,
+      "loss": 7.3977,
+      "step": 4000
+    },
+    {
+      "epoch": 1.8891687657430731,
+      "grad_norm": 6.854773044586182,
+      "learning_rate": 4.5e-05,
+      "loss": 7.3026,
+      "step": 4500
+    },
+    {
+      "epoch": 2.0990764063811924,
+      "grad_norm": 7.6127848625183105,
+      "learning_rate": 5e-05,
+      "loss": 7.1895,
+      "step": 5000
+    },
+    {
+      "epoch": 2.3089840470193117,
+      "grad_norm": 7.826136589050293,
+      "learning_rate": 5.500000000000001e-05,
+      "loss": 7.1128,
+      "step": 5500
+    },
+    {
+      "epoch": 2.5188916876574305,
+      "grad_norm": 8.711495399475098,
+      "learning_rate": 6e-05,
+      "loss": 7.0718,
+      "step": 6000
+    },
+    {
+      "epoch": 2.72879932829555,
+      "grad_norm": 4.3706536293029785,
+      "learning_rate": 6.500000000000001e-05,
+      "loss": 7.0242,
+      "step": 6500
+    },
+    {
+      "epoch": 2.938706968933669,
+      "grad_norm": 7.588512420654297,
+      "learning_rate": 7e-05,
+      "loss": 6.9993,
+      "step": 7000
+    },
+    {
+      "epoch": 3.1486146095717884,
+      "grad_norm": 6.372596740722656,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 6.9168,
+      "step": 7500
+    },
+    {
+      "epoch": 3.3585222502099077,
+      "grad_norm": 5.352084159851074,
+      "learning_rate": 8e-05,
+      "loss": 6.8948,
+      "step": 8000
+    },
+    {
+      "epoch": 3.568429890848027,
+      "grad_norm": 4.005120277404785,
+      "learning_rate": 8.5e-05,
+      "loss": 6.8742,
+      "step": 8500
+    },
+    {
+      "epoch": 3.7783375314861463,
+      "grad_norm": 3.1480915546417236,
+      "learning_rate": 9e-05,
+      "loss": 6.8502,
+      "step": 9000
+    },
+    {
+      "epoch": 3.988245172124265,
+      "grad_norm": 2.2671101093292236,
+      "learning_rate": 9.5e-05,
+      "loss": 6.8378,
+      "step": 9500
+    },
+    {
+      "epoch": 4.198152812762385,
+      "grad_norm": 3.7820088863372803,
+      "learning_rate": 0.0001,
+      "loss": 6.741,
+      "step": 10000
+    },
+    {
+      "epoch": 4.198152812762385,
+      "eval_accuracy": 0.15047719747349675,
+      "eval_loss": 6.86174201965332,
+      "eval_runtime": 1.0657,
+      "eval_samples_per_second": 449.453,
+      "eval_steps_per_second": 4.692,
+      "step": 10000
+    },
+    {
+      "epoch": 4.408060453400504,
+      "grad_norm": 2.66556978225708,
+      "learning_rate": 9.899091826437942e-05,
+      "loss": 6.741,
+      "step": 10500
+    },
+    {
+      "epoch": 4.617968094038623,
+      "grad_norm": 3.0062286853790283,
+      "learning_rate": 9.798183652875883e-05,
+      "loss": 6.7367,
+      "step": 11000
+    },
+    {
+      "epoch": 4.827875734676742,
+      "grad_norm": 2.927842855453491,
+      "learning_rate": 9.697275479313825e-05,
+      "loss": 6.7229,
+      "step": 11500
+    },
+    {
+      "epoch": 5.037783375314861,
+      "grad_norm": 3.1912167072296143,
+      "learning_rate": 9.596367305751767e-05,
+      "loss": 6.6724,
+      "step": 12000
+    },
+    {
+      "epoch": 5.247691015952981,
+      "grad_norm": 3.3745853900909424,
+      "learning_rate": 9.495459132189707e-05,
+      "loss": 6.6049,
+      "step": 12500
+    },
+    {
+      "epoch": 5.4575986565911,
+      "grad_norm": 2.804530382156372,
+      "learning_rate": 9.394550958627649e-05,
+      "loss": 6.5968,
+      "step": 13000
+    },
+    {
+      "epoch": 5.667506297229219,
+      "grad_norm": 3.8129560947418213,
+      "learning_rate": 9.293642785065591e-05,
+      "loss": 6.5992,
+      "step": 13500
+    },
+    {
+      "epoch": 5.877413937867338,
+      "grad_norm": 2.3966245651245117,
+      "learning_rate": 9.192734611503532e-05,
+      "loss": 6.5829,
+      "step": 14000
+    },
+    {
+      "epoch": 6.087321578505458,
+      "grad_norm": 3.265836477279663,
+      "learning_rate": 9.091826437941474e-05,
+      "loss": 6.5406,
+      "step": 14500
+    },
+    {
+      "epoch": 6.297229219143577,
+      "grad_norm": 2.9976537227630615,
+      "learning_rate": 8.990918264379415e-05,
+      "loss": 6.4934,
+      "step": 15000
+    },
+    {
+      "epoch": 6.507136859781696,
+      "grad_norm": 1.8768919706344604,
+      "learning_rate": 8.890010090817356e-05,
+      "loss": 6.4939,
+      "step": 15500
+    },
+    {
+      "epoch": 6.717044500419815,
+      "grad_norm": 3.4282023906707764,
+      "learning_rate": 8.789101917255298e-05,
+      "loss": 6.4879,
+      "step": 16000
+    },
+    {
+      "epoch": 6.926952141057934,
+      "grad_norm": 3.0579543113708496,
+      "learning_rate": 8.68819374369324e-05,
+      "loss": 6.4843,
+      "step": 16500
+    },
+    {
+      "epoch": 7.136859781696054,
+      "grad_norm": 2.4434196949005127,
+      "learning_rate": 8.58728557013118e-05,
+      "loss": 6.4169,
+      "step": 17000
+    },
+    {
+      "epoch": 7.346767422334173,
+      "grad_norm": 2.620242118835449,
+      "learning_rate": 8.486377396569122e-05,
+      "loss": 6.3983,
+      "step": 17500
+    },
+    {
+      "epoch": 7.5566750629722925,
+      "grad_norm": 2.902372121810913,
+      "learning_rate": 8.385469223007064e-05,
+      "loss": 6.4067,
+      "step": 18000
+    },
+    {
+      "epoch": 7.766582703610411,
+      "grad_norm": 2.9418728351593018,
+      "learning_rate": 8.284561049445006e-05,
+      "loss": 6.4046,
+      "step": 18500
+    },
+    {
+      "epoch": 7.97649034424853,
+      "grad_norm": 3.1700100898742676,
+      "learning_rate": 8.183652875882947e-05,
+      "loss": 6.4043,
+      "step": 19000
+    },
+    {
+      "epoch": 8.18639798488665,
+      "grad_norm": 2.1963798999786377,
+      "learning_rate": 8.082744702320889e-05,
+      "loss": 6.3264,
+      "step": 19500
+    },
+    {
+      "epoch": 8.39630562552477,
+      "grad_norm": 2.7740824222564697,
+      "learning_rate": 7.98183652875883e-05,
+      "loss": 6.3291,
+      "step": 20000
+    },
+    {
+      "epoch": 8.39630562552477,
+      "eval_accuracy": 0.15233311821319342,
+      "eval_loss": 6.701448440551758,
+      "eval_runtime": 1.0499,
+      "eval_samples_per_second": 456.231,
+      "eval_steps_per_second": 4.762,
+      "step": 20000
+    },
+    {
+      "epoch": 8.606213266162888,
+      "grad_norm": 2.7811079025268555,
+      "learning_rate": 7.880928355196771e-05,
+      "loss": 6.3322,
+      "step": 20500
+    },
+    {
+      "epoch": 8.816120906801007,
+      "grad_norm": 2.257051706314087,
+      "learning_rate": 7.780020181634713e-05,
+      "loss": 6.3292,
+      "step": 21000
+    },
+    {
+      "epoch": 9.026028547439127,
+      "grad_norm": 2.737684726715088,
+      "learning_rate": 7.679112008072655e-05,
+      "loss": 6.3188,
+      "step": 21500
+    },
+    {
+      "epoch": 9.235936188077247,
+      "grad_norm": 3.8119022846221924,
+      "learning_rate": 7.578203834510595e-05,
+      "loss": 6.2548,
+      "step": 22000
+    },
+    {
+      "epoch": 9.445843828715365,
+      "grad_norm": 2.7535812854766846,
+      "learning_rate": 7.477295660948537e-05,
+      "loss": 6.2632,
+      "step": 22500
+    },
+    {
+      "epoch": 9.655751469353484,
+      "grad_norm": 5.819505214691162,
+      "learning_rate": 7.376387487386479e-05,
+      "loss": 6.2604,
+      "step": 23000
+    },
+    {
+      "epoch": 9.865659109991604,
+      "grad_norm": 3.6783864498138428,
+      "learning_rate": 7.27547931382442e-05,
+      "loss": 6.2687,
+      "step": 23500
+    },
+    {
+      "epoch": 10.075566750629722,
+      "grad_norm": 5.783772945404053,
+      "learning_rate": 7.174571140262362e-05,
+      "loss": 6.231,
+      "step": 24000
+    },
+    {
+      "epoch": 10.285474391267842,
+      "grad_norm": 2.723422050476074,
+      "learning_rate": 7.073662966700304e-05,
+      "loss": 6.2044,
+      "step": 24500
+    },
+    {
+      "epoch": 10.495382031905962,
+      "grad_norm": 3.333406448364258,
+      "learning_rate": 6.972754793138244e-05,
+      "loss": 6.205,
+      "step": 25000
+    },
+    {
+      "epoch": 10.705289672544081,
+      "grad_norm": 3.7053959369659424,
+      "learning_rate": 6.871846619576186e-05,
+      "loss": 6.2037,
+      "step": 25500
+    },
+    {
+      "epoch": 10.9151973131822,
+      "grad_norm": 3.2251012325286865,
+      "learning_rate": 6.770938446014128e-05,
+      "loss": 6.2136,
+      "step": 26000
+    },
+    {
+      "epoch": 11.125104953820319,
+      "grad_norm": 2.389143466949463,
+      "learning_rate": 6.670030272452068e-05,
+      "loss": 6.1707,
+      "step": 26500
+    },
+    {
+      "epoch": 11.335012594458439,
+      "grad_norm": 3.476702928543091,
+      "learning_rate": 6.56912209889001e-05,
+      "loss": 6.1547,
+      "step": 27000
+    },
+    {
+      "epoch": 11.544920235096557,
+      "grad_norm": 3.1944007873535156,
+      "learning_rate": 6.468213925327952e-05,
+      "loss": 6.1514,
+      "step": 27500
+    },
+    {
+      "epoch": 11.754827875734676,
+      "grad_norm": 3.1503257751464844,
+      "learning_rate": 6.367305751765893e-05,
+      "loss": 6.155,
+      "step": 28000
+    },
+    {
+      "epoch": 11.964735516372796,
+      "grad_norm": 2.4318435192108154,
+      "learning_rate": 6.266397578203835e-05,
+      "loss": 6.1554,
+      "step": 28500
+    },
+    {
+      "epoch": 12.174643157010916,
+      "grad_norm": 2.668941020965576,
+      "learning_rate": 6.165489404641777e-05,
+      "loss": 6.1185,
+      "step": 29000
+    },
+    {
+      "epoch": 12.384550797649034,
+      "grad_norm": 2.188319206237793,
+      "learning_rate": 6.064581231079718e-05,
+      "loss": 6.1029,
+      "step": 29500
+    },
+    {
+      "epoch": 12.594458438287154,
+      "grad_norm": 2.9347007274627686,
+      "learning_rate": 5.963673057517659e-05,
+      "loss": 6.1153,
+      "step": 30000
+    },
+    {
+      "epoch": 12.594458438287154,
+      "eval_accuracy": 0.1539580826594286,
+      "eval_loss": 6.706422805786133,
+      "eval_runtime": 1.033,
+      "eval_samples_per_second": 463.676,
+      "eval_steps_per_second": 4.84,
+      "step": 30000
+    },
+    {
+      "epoch": 12.804366078925273,
+      "grad_norm": 3.4366564750671387,
+      "learning_rate": 5.8627648839556e-05,
+      "loss": 6.1044,
+      "step": 30500
+    },
+    {
+      "epoch": 13.014273719563391,
+      "grad_norm": 3.167576313018799,
+      "learning_rate": 5.7618567103935416e-05,
+      "loss": 6.1106,
+      "step": 31000
+    },
+    {
+      "epoch": 13.224181360201511,
+      "grad_norm": 2.0107972621917725,
+      "learning_rate": 5.660948536831484e-05,
+      "loss": 6.0733,
+      "step": 31500
+    },
+    {
+      "epoch": 13.43408900083963,
+      "grad_norm": 4.030771255493164,
+      "learning_rate": 5.5600403632694254e-05,
+      "loss": 6.072,
+      "step": 32000
+    },
+    {
+      "epoch": 13.64399664147775,
+      "grad_norm": 3.8046839237213135,
+      "learning_rate": 5.4591321897073666e-05,
+      "loss": 6.0786,
+      "step": 32500
+    },
+    {
+      "epoch": 13.853904282115868,
+      "grad_norm": 2.889934778213501,
+      "learning_rate": 5.358224016145308e-05,
+      "loss": 6.0757,
+      "step": 33000
+    },
+    {
+      "epoch": 14.063811922753988,
+      "grad_norm": 4.883760929107666,
+      "learning_rate": 5.257315842583249e-05,
+      "loss": 6.0633,
+      "step": 33500
+    },
+    {
+      "epoch": 14.273719563392108,
+      "grad_norm": 0.8915508389472961,
+      "learning_rate": 5.15640766902119e-05,
+      "loss": 6.0436,
+      "step": 34000
+    },
+    {
+      "epoch": 14.483627204030226,
+      "grad_norm": 2.643712043762207,
+      "learning_rate": 5.055499495459133e-05,
+      "loss": 6.0413,
+      "step": 34500
+    },
+    {
+      "epoch": 14.693534844668346,
+      "grad_norm": 2.5184438228607178,
+      "learning_rate": 4.954591321897074e-05,
+      "loss": 6.0432,
+      "step": 35000
+    },
+    {
+      "epoch": 14.903442485306465,
+      "grad_norm": 2.847947835922241,
+      "learning_rate": 4.853683148335015e-05,
+      "loss": 6.0409,
+      "step": 35500
+    },
+    {
+      "epoch": 15.113350125944585,
+      "grad_norm": 1.147056221961975,
+      "learning_rate": 4.7527749747729566e-05,
+      "loss": 6.0247,
+      "step": 36000
+    },
+    {
+      "epoch": 15.323257766582703,
+      "grad_norm": 2.816908597946167,
+      "learning_rate": 4.6518668012108985e-05,
+      "loss": 6.0205,
+      "step": 36500
+    },
+    {
+      "epoch": 15.533165407220823,
+      "grad_norm": 3.443824052810669,
+      "learning_rate": 4.55095862764884e-05,
+      "loss": 6.0149,
+      "step": 37000
+    },
+    {
+      "epoch": 15.743073047858942,
+      "grad_norm": 2.6054351329803467,
+      "learning_rate": 4.450050454086781e-05,
+      "loss": 6.0135,
+      "step": 37500
+    },
+    {
+      "epoch": 15.95298068849706,
+      "grad_norm": 3.3810739517211914,
+      "learning_rate": 4.349142280524723e-05,
+      "loss": 6.0131,
+      "step": 38000
+    },
+    {
+      "epoch": 16.162888329135182,
+      "grad_norm": 2.275756597518921,
+      "learning_rate": 4.248234106962664e-05,
+      "loss": 5.9965,
+      "step": 38500
+    },
+    {
+      "epoch": 16.3727959697733,
+      "grad_norm": 2.3946337699890137,
+      "learning_rate": 4.147325933400605e-05,
+      "loss": 5.9914,
+      "step": 39000
+    },
+    {
+      "epoch": 16.582703610411418,
+      "grad_norm": 4.954773426055908,
+      "learning_rate": 4.046417759838547e-05,
+      "loss": 5.9909,
+      "step": 39500
+    },
+    {
+      "epoch": 16.79261125104954,
+      "grad_norm": 4.085302829742432,
+      "learning_rate": 3.9455095862764884e-05,
+      "loss": 5.9862,
+      "step": 40000
+    },
+    {
+      "epoch": 16.79261125104954,
+      "eval_accuracy": 0.1519849725243916,
+      "eval_loss": 7.014549255371094,
+      "eval_runtime": 1.0149,
+      "eval_samples_per_second": 471.967,
+      "eval_steps_per_second": 4.927,
+      "step": 40000
+    },
+    {
+      "epoch": 17.002518891687657,
+      "grad_norm": 3.3821780681610107,
+      "learning_rate": 3.84460141271443e-05,
+      "loss": 5.9872,
+      "step": 40500
+    },
+    {
+      "epoch": 17.212426532325775,
+      "grad_norm": 1.8607794046401978,
+      "learning_rate": 3.7436932391523715e-05,
+      "loss": 5.9711,
+      "step": 41000
+    },
+    {
+      "epoch": 17.422334172963897,
+      "grad_norm": 2.3415207862854004,
+      "learning_rate": 3.642785065590313e-05,
+      "loss": 5.9652,
+      "step": 41500
+    },
+    {
+      "epoch": 17.632241813602015,
+      "grad_norm": 2.3748843669891357,
+      "learning_rate": 3.541876892028255e-05,
+      "loss": 5.97,
+      "step": 42000
+    },
+    {
+      "epoch": 17.842149454240133,
+      "grad_norm": 1.7207521200180054,
+      "learning_rate": 3.440968718466196e-05,
+      "loss": 5.9688,
+      "step": 42500
+    },
+    {
+      "epoch": 18.052057094878254,
+      "grad_norm": 3.5838589668273926,
+      "learning_rate": 3.340060544904137e-05,
+      "loss": 5.9576,
+      "step": 43000
+    },
+    {
+      "epoch": 18.261964735516372,
+      "grad_norm": 1.1076291799545288,
+      "learning_rate": 3.239152371342079e-05,
+      "loss": 5.9478,
+      "step": 43500
+    },
+    {
+      "epoch": 18.471872376154494,
+      "grad_norm": 2.307013750076294,
+      "learning_rate": 3.13824419778002e-05,
+      "loss": 5.9504,
+      "step": 44000
+    },
+    {
+      "epoch": 18.68178001679261,
+      "grad_norm": 1.877256989479065,
+      "learning_rate": 3.0373360242179615e-05,
+      "loss": 5.95,
+      "step": 44500
+    },
+    {
+      "epoch": 18.89168765743073,
+      "grad_norm": 2.667221784591675,
+      "learning_rate": 2.9364278506559034e-05,
+      "loss": 5.9522,
+      "step": 45000
+    },
+    {
+      "epoch": 19.10159529806885,
+      "grad_norm": 1.781567096710205,
+      "learning_rate": 2.8355196770938446e-05,
+      "loss": 5.9428,
+      "step": 45500
+    },
+    {
+      "epoch": 19.31150293870697,
+      "grad_norm": 2.600219488143921,
+      "learning_rate": 2.734611503531786e-05,
+      "loss": 5.9273,
+      "step": 46000
+    },
+    {
+      "epoch": 19.521410579345087,
+      "grad_norm": 3.2653560638427734,
+      "learning_rate": 2.6337033299697278e-05,
+      "loss": 5.9327,
+      "step": 46500
+    },
+    {
+      "epoch": 19.73131821998321,
+      "grad_norm": 0.8226433992385864,
+      "learning_rate": 2.532795156407669e-05,
+      "loss": 5.9303,
+      "step": 47000
+    },
+    {
+      "epoch": 19.941225860621326,
+      "grad_norm": 2.6203231811523438,
+      "learning_rate": 2.4318869828456106e-05,
+      "loss": 5.9259,
+      "step": 47500
+    },
+    {
+      "epoch": 20.151133501259444,
+      "grad_norm": 3.7636687755584717,
+      "learning_rate": 2.330978809283552e-05,
+      "loss": 5.9187,
+      "step": 48000
+    },
+    {
+      "epoch": 20.361041141897566,
+      "grad_norm": 2.3655290603637695,
+      "learning_rate": 2.2300706357214934e-05,
+      "loss": 5.9176,
+      "step": 48500
+    },
+    {
+      "epoch": 20.570948782535684,
+      "grad_norm": 1.55186927318573,
+      "learning_rate": 2.129162462159435e-05,
+      "loss": 5.9103,
+      "step": 49000
+    },
+    {
+      "epoch": 20.7808564231738,
+      "grad_norm": 4.357689380645752,
+      "learning_rate": 2.0282542885973765e-05,
+      "loss": 5.9109,
+      "step": 49500
+    },
+    {
+      "epoch": 20.990764063811923,
+      "grad_norm": 3.323455333709717,
+      "learning_rate": 1.927346115035318e-05,
+      "loss": 5.9088,
+      "step": 50000
+    },
+    {
+      "epoch": 20.990764063811923,
+      "eval_accuracy": 0.15313875571256344,
+      "eval_loss": 7.005898952484131,
+      "eval_runtime": 1.0145,
+      "eval_samples_per_second": 472.149,
+      "eval_steps_per_second": 4.928,
+      "step": 50000
+    },
+    {
+      "epoch": 21.20067170445004,
+      "grad_norm": 0.8441404104232788,
+      "learning_rate": 1.8264379414732593e-05,
+      "loss": 5.9017,
+      "step": 50500
+    },
+    {
+      "epoch": 21.410579345088163,
+      "grad_norm": 2.293954372406006,
+      "learning_rate": 1.725529767911201e-05,
+      "loss": 5.9007,
+      "step": 51000
+    },
+    {
+      "epoch": 21.62048698572628,
+      "grad_norm": 2.9091224670410156,
+      "learning_rate": 1.6246215943491424e-05,
+      "loss": 5.9001,
+      "step": 51500
+    },
+    {
+      "epoch": 21.8303946263644,
+      "grad_norm": 3.189171075820923,
+      "learning_rate": 1.5237134207870837e-05,
+      "loss": 5.8969,
+      "step": 52000
+    },
+    {
+      "epoch": 22.04030226700252,
+      "grad_norm": 0.8301982283592224,
+      "learning_rate": 1.4228052472250252e-05,
+      "loss": 5.8956,
+      "step": 52500
+    },
+    {
+      "epoch": 22.250209907640638,
+      "grad_norm": 3.955568552017212,
+      "learning_rate": 1.3218970736629668e-05,
+      "loss": 5.8826,
+      "step": 53000
+    },
+    {
+      "epoch": 22.460117548278756,
+      "grad_norm": 1.694311261177063,
+      "learning_rate": 1.2209889001009082e-05,
+      "loss": 5.8871,
+      "step": 53500
+    },
+    {
+      "epoch": 22.670025188916878,
+      "grad_norm": 1.507860541343689,
+      "learning_rate": 1.1200807265388496e-05,
+      "loss": 5.89,
+      "step": 54000
+    },
+    {
+      "epoch": 22.879932829554996,
+      "grad_norm": 0.8334353566169739,
+      "learning_rate": 1.0191725529767912e-05,
+      "loss": 5.8817,
+      "step": 54500
+    },
+    {
+      "epoch": 23.089840470193113,
+      "grad_norm": 0.8316900134086609,
+      "learning_rate": 9.182643794147326e-06,
+      "loss": 5.8804,
+      "step": 55000
+    },
+    {
+      "epoch": 23.299748110831235,
+      "grad_norm": 2.653756618499756,
+      "learning_rate": 8.173562058526741e-06,
+      "loss": 5.8749,
+      "step": 55500
+    },
+    {
+      "epoch": 23.509655751469353,
+      "grad_norm": 3.7355833053588867,
+      "learning_rate": 7.164480322906156e-06,
+      "loss": 5.8769,
+      "step": 56000
+    },
+    {
+      "epoch": 23.719563392107474,
+      "grad_norm": 0.8122338652610779,
+      "learning_rate": 6.155398587285571e-06,
+      "loss": 5.8699,
+      "step": 56500
+    },
+    {
+      "epoch": 23.929471032745592,
+      "grad_norm": 0.8836779594421387,
+      "learning_rate": 5.146316851664986e-06,
+      "loss": 5.8733,
+      "step": 57000
+    },
+    {
+      "epoch": 24.13937867338371,
+      "grad_norm": 3.144986152648926,
+      "learning_rate": 4.1372351160444e-06,
+      "loss": 5.8685,
+      "step": 57500
+    },
+    {
+      "epoch": 24.349286314021832,
+      "grad_norm": 3.1895604133605957,
+      "learning_rate": 3.1281533804238145e-06,
+      "loss": 5.8702,
+      "step": 58000
+    },
+    {
+      "epoch": 24.55919395465995,
+      "grad_norm": 0.9001981019973755,
+      "learning_rate": 2.1190716448032293e-06,
+      "loss": 5.8651,
+      "step": 58500
+    },
+    {
+      "epoch": 24.769101595298068,
+      "grad_norm": 1.6653518676757812,
+      "learning_rate": 1.109989909182644e-06,
+      "loss": 5.8625,
+      "step": 59000
+    },
+    {
+      "epoch": 24.97900923593619,
+      "grad_norm": 1.8964729309082031,
+      "learning_rate": 1.0090817356205853e-07,
+      "loss": 5.8636,
+      "step": 59500
+    },
+    {
+      "epoch": 25.0,
+      "step": 59550,
+      "total_flos": 1.5151126801503744e+18,
+      "train_loss": 6.378149575490495,
+      "train_runtime": 24807.4865,
+      "train_samples_per_second": 230.413,
+      "train_steps_per_second": 2.4
+    }
+  ],
+  "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": 1.5151126801503744e+18,
+  "train_batch_size": 96,
+  "trial_name": null,
+  "trial_params": null
+}