End of training

README.md

@@ -2,11 +2,23 @@
 library_name: transformers
 tags:
 - generated_from_trainer
+datasets:
+- Hartunka/processed_wikitext-103-raw-v1-rand-5_v2
 metrics:
 - accuracy
 model-index:
 - name: distilbert_rand_5_v2
-  results: []
+  results:
+  - task:
+      name: Masked Language Modeling
+      type: fill-mask
+    dataset:
+      name: Hartunka/processed_wikitext-103-raw-v1-rand-5_v2
+      type: Hartunka/processed_wikitext-103-raw-v1-rand-5_v2
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.15330281968695258
 ---
 
 <!-- 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. -->
 
 # distilbert_rand_5_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-5_v2 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 10.1356
-- Accuracy: 0.1513
+- Loss: 7.6888
+- Accuracy: 0.1533
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 25.0,
+    "eval_accuracy": 0.15330281968695258,
+    "eval_loss": 7.68883752822876,
+    "eval_runtime": 0.8084,
+    "eval_samples": 479,
+    "eval_samples_per_second": 592.541,
+    "eval_steps_per_second": 6.185,
+    "perplexity": 2183.8344402144076,
+    "total_flos": 7.68154255975296e+17,
+    "train_loss": 6.938646208973716,
+    "train_runtime": 16943.8632,
+    "train_samples": 228639,
+    "train_samples_per_second": 337.348,
+    "train_steps_per_second": 3.515
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 25.0,
+    "eval_accuracy": 0.15330281968695258,
+    "eval_loss": 7.68883752822876,
+    "eval_runtime": 0.8084,
+    "eval_samples": 479,
+    "eval_samples_per_second": 592.541,
+    "eval_steps_per_second": 6.185,
+    "perplexity": 2183.8344402144076
+}

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 25.0,
+    "total_flos": 7.68154255975296e+17,
+    "train_loss": 6.938646208973716,
+    "train_runtime": 16943.8632,
+    "train_samples": 228639,
+    "train_samples_per_second": 337.348,
+    "train_steps_per_second": 3.515
+}

trainer_state.json

@@ -0,0 +1,921 @@
+{
+  "best_global_step": 10000,
+  "best_metric": 7.672569274902344,
+  "best_model_checkpoint": "distilbert_rand_5_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": 2.696444272994995,
+      "learning_rate": 5e-06,
+      "loss": 10.9028,
+      "step": 500
+    },
+    {
+      "epoch": 0.41981528127623846,
+      "grad_norm": 1.039544701576233,
+      "learning_rate": 1e-05,
+      "loss": 9.1806,
+      "step": 1000
+    },
+    {
+      "epoch": 0.6297229219143576,
+      "grad_norm": 2.070500612258911,
+      "learning_rate": 1.5e-05,
+      "loss": 8.6545,
+      "step": 1500
+    },
+    {
+      "epoch": 0.8396305625524769,
+      "grad_norm": 3.3437089920043945,
+      "learning_rate": 2e-05,
+      "loss": 8.4956,
+      "step": 2000
+    },
+    {
+      "epoch": 1.0495382031905962,
+      "grad_norm": 3.614703893661499,
+      "learning_rate": 2.5e-05,
+      "loss": 8.36,
+      "step": 2500
+    },
+    {
+      "epoch": 1.2594458438287153,
+      "grad_norm": 2.2285263538360596,
+      "learning_rate": 3e-05,
+      "loss": 8.2504,
+      "step": 3000
+    },
+    {
+      "epoch": 1.4693534844668346,
+      "grad_norm": 2.225768566131592,
+      "learning_rate": 3.5e-05,
+      "loss": 8.1561,
+      "step": 3500
+    },
+    {
+      "epoch": 1.6792611251049538,
+      "grad_norm": 1.565374493598938,
+      "learning_rate": 4e-05,
+      "loss": 8.0871,
+      "step": 4000
+    },
+    {
+      "epoch": 1.8891687657430731,
+      "grad_norm": 2.018296241760254,
+      "learning_rate": 4.5e-05,
+      "loss": 8.0323,
+      "step": 4500
+    },
+    {
+      "epoch": 2.0990764063811924,
+      "grad_norm": 2.5747921466827393,
+      "learning_rate": 5e-05,
+      "loss": 7.9835,
+      "step": 5000
+    },
+    {
+      "epoch": 2.3089840470193117,
+      "grad_norm": 3.364144802093506,
+      "learning_rate": 5.500000000000001e-05,
+      "loss": 7.944,
+      "step": 5500
+    },
+    {
+      "epoch": 2.5188916876574305,
+      "grad_norm": 1.9467853307724,
+      "learning_rate": 6e-05,
+      "loss": 7.9083,
+      "step": 6000
+    },
+    {
+      "epoch": 2.72879932829555,
+      "grad_norm": 2.5052247047424316,
+      "learning_rate": 6.500000000000001e-05,
+      "loss": 7.8749,
+      "step": 6500
+    },
+    {
+      "epoch": 2.938706968933669,
+      "grad_norm": 2.1754403114318848,
+      "learning_rate": 7e-05,
+      "loss": 7.8449,
+      "step": 7000
+    },
+    {
+      "epoch": 3.1486146095717884,
+      "grad_norm": 2.9289937019348145,
+      "learning_rate": 7.500000000000001e-05,
+      "loss": 7.817,
+      "step": 7500
+    },
+    {
+      "epoch": 3.3585222502099077,
+      "grad_norm": 1.858054757118225,
+      "learning_rate": 8e-05,
+      "loss": 7.7941,
+      "step": 8000
+    },
+    {
+      "epoch": 3.568429890848027,
+      "grad_norm": 1.465584635734558,
+      "learning_rate": 8.5e-05,
+      "loss": 7.7788,
+      "step": 8500
+    },
+    {
+      "epoch": 3.7783375314861463,
+      "grad_norm": 1.4048516750335693,
+      "learning_rate": 9e-05,
+      "loss": 7.7546,
+      "step": 9000
+    },
+    {
+      "epoch": 3.988245172124265,
+      "grad_norm": 3.0935544967651367,
+      "learning_rate": 9.5e-05,
+      "loss": 7.7381,
+      "step": 9500
+    },
+    {
+      "epoch": 4.198152812762385,
+      "grad_norm": 2.5529305934906006,
+      "learning_rate": 0.0001,
+      "loss": 7.6942,
+      "step": 10000
+    },
+    {
+      "epoch": 4.198152812762385,
+      "eval_accuracy": 0.15103369598486324,
+      "eval_loss": 7.672569274902344,
+      "eval_runtime": 0.8828,
+      "eval_samples_per_second": 542.621,
+      "eval_steps_per_second": 5.664,
+      "step": 10000
+    },
+    {
+      "epoch": 4.408060453400504,
+      "grad_norm": 2.0559029579162598,
+      "learning_rate": 9.899091826437942e-05,
+      "loss": 7.687,
+      "step": 10500
+    },
+    {
+      "epoch": 4.617968094038623,
+      "grad_norm": 1.7670162916183472,
+      "learning_rate": 9.798183652875883e-05,
+      "loss": 7.6675,
+      "step": 11000
+    },
+    {
+      "epoch": 4.827875734676742,
+      "grad_norm": 2.2571051120758057,
+      "learning_rate": 9.697275479313825e-05,
+      "loss": 7.6653,
+      "step": 11500
+    },
+    {
+      "epoch": 5.037783375314861,
+      "grad_norm": 2.5738790035247803,
+      "learning_rate": 9.596367305751767e-05,
+      "loss": 7.6399,
+      "step": 12000
+    },
+    {
+      "epoch": 5.247691015952981,
+      "grad_norm": 2.763585090637207,
+      "learning_rate": 9.495459132189707e-05,
+      "loss": 7.5871,
+      "step": 12500
+    },
+    {
+      "epoch": 5.4575986565911,
+      "grad_norm": 2.1902973651885986,
+      "learning_rate": 9.394550958627649e-05,
+      "loss": 7.5842,
+      "step": 13000
+    },
+    {
+      "epoch": 5.667506297229219,
+      "grad_norm": 1.5051831007003784,
+      "learning_rate": 9.293642785065591e-05,
+      "loss": 7.5886,
+      "step": 13500
+    },
+    {
+      "epoch": 5.877413937867338,
+      "grad_norm": 2.7956175804138184,
+      "learning_rate": 9.192734611503532e-05,
+      "loss": 7.579,
+      "step": 14000
+    },
+    {
+      "epoch": 6.087321578505458,
+      "grad_norm": 2.817504644393921,
+      "learning_rate": 9.091826437941474e-05,
+      "loss": 7.5472,
+      "step": 14500
+    },
+    {
+      "epoch": 6.297229219143577,
+      "grad_norm": 3.2463319301605225,
+      "learning_rate": 8.990918264379415e-05,
+      "loss": 7.5188,
+      "step": 15000
+    },
+    {
+      "epoch": 6.507136859781696,
+      "grad_norm": 3.793332099914551,
+      "learning_rate": 8.890010090817356e-05,
+      "loss": 7.5163,
+      "step": 15500
+    },
+    {
+      "epoch": 6.717044500419815,
+      "grad_norm": 3.2017998695373535,
+      "learning_rate": 8.789101917255298e-05,
+      "loss": 7.5196,
+      "step": 16000
+    },
+    {
+      "epoch": 6.926952141057934,
+      "grad_norm": 2.842729091644287,
+      "learning_rate": 8.68819374369324e-05,
+      "loss": 7.5181,
+      "step": 16500
+    },
+    {
+      "epoch": 7.136859781696054,
+      "grad_norm": 4.094554901123047,
+      "learning_rate": 8.58728557013118e-05,
+      "loss": 7.4687,
+      "step": 17000
+    },
+    {
+      "epoch": 7.346767422334173,
+      "grad_norm": 2.4635562896728516,
+      "learning_rate": 8.486377396569122e-05,
+      "loss": 7.449,
+      "step": 17500
+    },
+    {
+      "epoch": 7.5566750629722925,
+      "grad_norm": 4.259366512298584,
+      "learning_rate": 8.385469223007064e-05,
+      "loss": 7.4578,
+      "step": 18000
+    },
+    {
+      "epoch": 7.766582703610411,
+      "grad_norm": 2.1686573028564453,
+      "learning_rate": 8.284561049445006e-05,
+      "loss": 7.4623,
+      "step": 18500
+    },
+    {
+      "epoch": 7.97649034424853,
+      "grad_norm": 2.079665184020996,
+      "learning_rate": 8.183652875882947e-05,
+      "loss": 7.4613,
+      "step": 19000
+    },
+    {
+      "epoch": 8.18639798488665,
+      "grad_norm": 3.346262216567993,
+      "learning_rate": 8.082744702320889e-05,
+      "loss": 7.3801,
+      "step": 19500
+    },
+    {
+      "epoch": 8.39630562552477,
+      "grad_norm": 2.8092551231384277,
+      "learning_rate": 7.98183652875883e-05,
+      "loss": 7.3808,
+      "step": 20000
+    },
+    {
+      "epoch": 8.39630562552477,
+      "eval_accuracy": 0.15191223098291198,
+      "eval_loss": 7.7572832107543945,
+      "eval_runtime": 0.8244,
+      "eval_samples_per_second": 581.007,
+      "eval_steps_per_second": 6.065,
+      "step": 20000
+    },
+    {
+      "epoch": 8.606213266162888,
+      "grad_norm": 2.968716621398926,
+      "learning_rate": 7.880928355196771e-05,
+      "loss": 7.3801,
+      "step": 20500
+    },
+    {
+      "epoch": 8.816120906801007,
+      "grad_norm": 3.2606894969940186,
+      "learning_rate": 7.780020181634713e-05,
+      "loss": 7.3957,
+      "step": 21000
+    },
+    {
+      "epoch": 9.026028547439127,
+      "grad_norm": 3.1347687244415283,
+      "learning_rate": 7.679112008072655e-05,
+      "loss": 7.3689,
+      "step": 21500
+    },
+    {
+      "epoch": 9.235936188077247,
+      "grad_norm": 3.8049638271331787,
+      "learning_rate": 7.578203834510595e-05,
+      "loss": 7.2589,
+      "step": 22000
+    },
+    {
+      "epoch": 9.445843828715365,
+      "grad_norm": 3.362259864807129,
+      "learning_rate": 7.477295660948537e-05,
+      "loss": 7.2697,
+      "step": 22500
+    },
+    {
+      "epoch": 9.655751469353484,
+      "grad_norm": 3.0395684242248535,
+      "learning_rate": 7.376387487386479e-05,
+      "loss": 7.2867,
+      "step": 23000
+    },
+    {
+      "epoch": 9.865659109991604,
+      "grad_norm": 3.318899154663086,
+      "learning_rate": 7.27547931382442e-05,
+      "loss": 7.2852,
+      "step": 23500
+    },
+    {
+      "epoch": 10.075566750629722,
+      "grad_norm": 4.256964206695557,
+      "learning_rate": 7.174571140262362e-05,
+      "loss": 7.2126,
+      "step": 24000
+    },
+    {
+      "epoch": 10.285474391267842,
+      "grad_norm": 4.575404644012451,
+      "learning_rate": 7.073662966700304e-05,
+      "loss": 7.1018,
+      "step": 24500
+    },
+    {
+      "epoch": 10.495382031905962,
+      "grad_norm": 4.5876569747924805,
+      "learning_rate": 6.972754793138244e-05,
+      "loss": 7.1135,
+      "step": 25000
+    },
+    {
+      "epoch": 10.705289672544081,
+      "grad_norm": 4.371004104614258,
+      "learning_rate": 6.871846619576186e-05,
+      "loss": 7.1284,
+      "step": 25500
+    },
+    {
+      "epoch": 10.9151973131822,
+      "grad_norm": 3.420358896255493,
+      "learning_rate": 6.770938446014128e-05,
+      "loss": 7.1428,
+      "step": 26000
+    },
+    {
+      "epoch": 11.125104953820319,
+      "grad_norm": 5.677443504333496,
+      "learning_rate": 6.670030272452068e-05,
+      "loss": 6.9964,
+      "step": 26500
+    },
+    {
+      "epoch": 11.335012594458439,
+      "grad_norm": 5.363876819610596,
+      "learning_rate": 6.56912209889001e-05,
+      "loss": 6.9201,
+      "step": 27000
+    },
+    {
+      "epoch": 11.544920235096557,
+      "grad_norm": 4.761507511138916,
+      "learning_rate": 6.468213925327952e-05,
+      "loss": 6.9377,
+      "step": 27500
+    },
+    {
+      "epoch": 11.754827875734676,
+      "grad_norm": 7.46794319152832,
+      "learning_rate": 6.367305751765893e-05,
+      "loss": 6.9471,
+      "step": 28000
+    },
+    {
+      "epoch": 11.964735516372796,
+      "grad_norm": 5.856936931610107,
+      "learning_rate": 6.266397578203835e-05,
+      "loss": 6.9659,
+      "step": 28500
+    },
+    {
+      "epoch": 12.174643157010916,
+      "grad_norm": 5.841556072235107,
+      "learning_rate": 6.165489404641777e-05,
+      "loss": 6.7691,
+      "step": 29000
+    },
+    {
+      "epoch": 12.384550797649034,
+      "grad_norm": 6.575760841369629,
+      "learning_rate": 6.064581231079718e-05,
+      "loss": 6.7455,
+      "step": 29500
+    },
+    {
+      "epoch": 12.594458438287154,
+      "grad_norm": 5.908180236816406,
+      "learning_rate": 5.963673057517659e-05,
+      "loss": 6.7779,
+      "step": 30000
+    },
+    {
+      "epoch": 12.594458438287154,
+      "eval_accuracy": 0.15451772672580238,
+      "eval_loss": 8.576528549194336,
+      "eval_runtime": 0.815,
+      "eval_samples_per_second": 587.759,
+      "eval_steps_per_second": 6.135,
+      "step": 30000
+    },
+    {
+      "epoch": 12.804366078925273,
+      "grad_norm": 6.440098762512207,
+      "learning_rate": 5.8627648839556e-05,
+      "loss": 6.7864,
+      "step": 30500
+    },
+    {
+      "epoch": 13.014273719563391,
+      "grad_norm": 7.687087535858154,
+      "learning_rate": 5.7618567103935416e-05,
+      "loss": 6.7867,
+      "step": 31000
+    },
+    {
+      "epoch": 13.224181360201511,
+      "grad_norm": 5.808923244476318,
+      "learning_rate": 5.660948536831484e-05,
+      "loss": 6.5896,
+      "step": 31500
+    },
+    {
+      "epoch": 13.43408900083963,
+      "grad_norm": 7.567235469818115,
+      "learning_rate": 5.5600403632694254e-05,
+      "loss": 6.6144,
+      "step": 32000
+    },
+    {
+      "epoch": 13.64399664147775,
+      "grad_norm": 6.806401252746582,
+      "learning_rate": 5.4591321897073666e-05,
+      "loss": 6.6406,
+      "step": 32500
+    },
+    {
+      "epoch": 13.853904282115868,
+      "grad_norm": 6.6164093017578125,
+      "learning_rate": 5.358224016145308e-05,
+      "loss": 6.6441,
+      "step": 33000
+    },
+    {
+      "epoch": 14.063811922753988,
+      "grad_norm": 7.657918453216553,
+      "learning_rate": 5.257315842583249e-05,
+      "loss": 6.5901,
+      "step": 33500
+    },
+    {
+      "epoch": 14.273719563392108,
+      "grad_norm": 6.972758769989014,
+      "learning_rate": 5.15640766902119e-05,
+      "loss": 6.4731,
+      "step": 34000
+    },
+    {
+      "epoch": 14.483627204030226,
+      "grad_norm": 5.738254547119141,
+      "learning_rate": 5.055499495459133e-05,
+      "loss": 6.4945,
+      "step": 34500
+    },
+    {
+      "epoch": 14.693534844668346,
+      "grad_norm": 6.05341100692749,
+      "learning_rate": 4.954591321897074e-05,
+      "loss": 6.5061,
+      "step": 35000
+    },
+    {
+      "epoch": 14.903442485306465,
+      "grad_norm": 8.559845924377441,
+      "learning_rate": 4.853683148335015e-05,
+      "loss": 6.5185,
+      "step": 35500
+    },
+    {
+      "epoch": 15.113350125944585,
+      "grad_norm": 7.044485569000244,
+      "learning_rate": 4.7527749747729566e-05,
+      "loss": 6.4371,
+      "step": 36000
+    },
+    {
+      "epoch": 15.323257766582703,
+      "grad_norm": 6.021819591522217,
+      "learning_rate": 4.6518668012108985e-05,
+      "loss": 6.3764,
+      "step": 36500
+    },
+    {
+      "epoch": 15.533165407220823,
+      "grad_norm": 5.949471473693848,
+      "learning_rate": 4.55095862764884e-05,
+      "loss": 6.3897,
+      "step": 37000
+    },
+    {
+      "epoch": 15.743073047858942,
+      "grad_norm": 5.850702285766602,
+      "learning_rate": 4.450050454086781e-05,
+      "loss": 6.4104,
+      "step": 37500
+    },
+    {
+      "epoch": 15.95298068849706,
+      "grad_norm": 6.750476360321045,
+      "learning_rate": 4.349142280524723e-05,
+      "loss": 6.4191,
+      "step": 38000
+    },
+    {
+      "epoch": 16.162888329135182,
+      "grad_norm": 8.018487930297852,
+      "learning_rate": 4.248234106962664e-05,
+      "loss": 6.3171,
+      "step": 38500
+    },
+    {
+      "epoch": 16.3727959697733,
+      "grad_norm": 7.288136005401611,
+      "learning_rate": 4.147325933400605e-05,
+      "loss": 6.3088,
+      "step": 39000
+    },
+    {
+      "epoch": 16.582703610411418,
+      "grad_norm": 7.808495998382568,
+      "learning_rate": 4.046417759838547e-05,
+      "loss": 6.3087,
+      "step": 39500
+    },
+    {
+      "epoch": 16.79261125104954,
+      "grad_norm": 6.476433277130127,
+      "learning_rate": 3.9455095862764884e-05,
+      "loss": 6.3211,
+      "step": 40000
+    },
+    {
+      "epoch": 16.79261125104954,
+      "eval_accuracy": 0.15220926320511383,
+      "eval_loss": 9.262114524841309,
+      "eval_runtime": 0.8065,
+      "eval_samples_per_second": 593.915,
+      "eval_steps_per_second": 6.2,
+      "step": 40000
+    },
+    {
+      "epoch": 17.002518891687657,
+      "grad_norm": 5.892916202545166,
+      "learning_rate": 3.84460141271443e-05,
+      "loss": 6.3259,
+      "step": 40500
+    },
+    {
+      "epoch": 17.212426532325775,
+      "grad_norm": 6.401475429534912,
+      "learning_rate": 3.7436932391523715e-05,
+      "loss": 6.2278,
+      "step": 41000
+    },
+    {
+      "epoch": 17.422334172963897,
+      "grad_norm": 8.905499458312988,
+      "learning_rate": 3.642785065590313e-05,
+      "loss": 6.2357,
+      "step": 41500
+    },
+    {
+      "epoch": 17.632241813602015,
+      "grad_norm": 8.250408172607422,
+      "learning_rate": 3.541876892028255e-05,
+      "loss": 6.2492,
+      "step": 42000
+    },
+    {
+      "epoch": 17.842149454240133,
+      "grad_norm": 6.09629487991333,
+      "learning_rate": 3.440968718466196e-05,
+      "loss": 6.2553,
+      "step": 42500
+    },
+    {
+      "epoch": 18.052057094878254,
+      "grad_norm": 5.08654260635376,
+      "learning_rate": 3.340060544904137e-05,
+      "loss": 6.2331,
+      "step": 43000
+    },
+    {
+      "epoch": 18.261964735516372,
+      "grad_norm": 7.321967124938965,
+      "learning_rate": 3.239152371342079e-05,
+      "loss": 6.1845,
+      "step": 43500
+    },
+    {
+      "epoch": 18.471872376154494,
+      "grad_norm": 6.764617443084717,
+      "learning_rate": 3.13824419778002e-05,
+      "loss": 6.1885,
+      "step": 44000
+    },
+    {
+      "epoch": 18.68178001679261,
+      "grad_norm": 5.396496772766113,
+      "learning_rate": 3.0373360242179615e-05,
+      "loss": 6.2027,
+      "step": 44500
+    },
+    {
+      "epoch": 18.89168765743073,
+      "grad_norm": 8.324110984802246,
+      "learning_rate": 2.9364278506559034e-05,
+      "loss": 6.2084,
+      "step": 45000
+    },
+    {
+      "epoch": 19.10159529806885,
+      "grad_norm": 5.621279716491699,
+      "learning_rate": 2.8355196770938446e-05,
+      "loss": 6.168,
+      "step": 45500
+    },
+    {
+      "epoch": 19.31150293870697,
+      "grad_norm": 6.803528785705566,
+      "learning_rate": 2.734611503531786e-05,
+      "loss": 6.1483,
+      "step": 46000
+    },
+    {
+      "epoch": 19.521410579345087,
+      "grad_norm": 6.71403169631958,
+      "learning_rate": 2.6337033299697278e-05,
+      "loss": 6.1525,
+      "step": 46500
+    },
+    {
+      "epoch": 19.73131821998321,
+      "grad_norm": 6.618786811828613,
+      "learning_rate": 2.532795156407669e-05,
+      "loss": 6.1517,
+      "step": 47000
+    },
+    {
+      "epoch": 19.941225860621326,
+      "grad_norm": 4.139350891113281,
+      "learning_rate": 2.4318869828456106e-05,
+      "loss": 6.152,
+      "step": 47500
+    },
+    {
+      "epoch": 20.151133501259444,
+      "grad_norm": 6.736079216003418,
+      "learning_rate": 2.330978809283552e-05,
+      "loss": 6.1325,
+      "step": 48000
+    },
+    {
+      "epoch": 20.361041141897566,
+      "grad_norm": 6.890057563781738,
+      "learning_rate": 2.2300706357214934e-05,
+      "loss": 6.1187,
+      "step": 48500
+    },
+    {
+      "epoch": 20.570948782535684,
+      "grad_norm": 4.702675819396973,
+      "learning_rate": 2.129162462159435e-05,
+      "loss": 6.1176,
+      "step": 49000
+    },
+    {
+      "epoch": 20.7808564231738,
+      "grad_norm": 6.187152862548828,
+      "learning_rate": 2.0282542885973765e-05,
+      "loss": 6.1202,
+      "step": 49500
+    },
+    {
+      "epoch": 20.990764063811923,
+      "grad_norm": 6.398833274841309,
+      "learning_rate": 1.927346115035318e-05,
+      "loss": 6.1233,
+      "step": 50000
+    },
+    {
+      "epoch": 20.990764063811923,
+      "eval_accuracy": 0.1512602685956206,
+      "eval_loss": 10.135626792907715,
+      "eval_runtime": 0.8124,
+      "eval_samples_per_second": 589.6,
+      "eval_steps_per_second": 6.154,
+      "step": 50000
+    },
+    {
+      "epoch": 21.20067170445004,
+      "grad_norm": 9.812323570251465,
+      "learning_rate": 1.8264379414732593e-05,
+      "loss": 6.0937,
+      "step": 50500
+    },
+    {
+      "epoch": 21.410579345088163,
+      "grad_norm": 6.78523063659668,
+      "learning_rate": 1.725529767911201e-05,
+      "loss": 6.0964,
+      "step": 51000
+    },
+    {
+      "epoch": 21.62048698572628,
+      "grad_norm": 9.46012020111084,
+      "learning_rate": 1.6246215943491424e-05,
+      "loss": 6.0932,
+      "step": 51500
+    },
+    {
+      "epoch": 21.8303946263644,
+      "grad_norm": 5.019435882568359,
+      "learning_rate": 1.5237134207870837e-05,
+      "loss": 6.0954,
+      "step": 52000
+    },
+    {
+      "epoch": 22.04030226700252,
+      "grad_norm": 5.651519298553467,
+      "learning_rate": 1.4228052472250252e-05,
+      "loss": 6.0899,
+      "step": 52500
+    },
+    {
+      "epoch": 22.250209907640638,
+      "grad_norm": 3.4696178436279297,
+      "learning_rate": 1.3218970736629668e-05,
+      "loss": 6.0678,
+      "step": 53000
+    },
+    {
+      "epoch": 22.460117548278756,
+      "grad_norm": 5.378692626953125,
+      "learning_rate": 1.2209889001009082e-05,
+      "loss": 6.0668,
+      "step": 53500
+    },
+    {
+      "epoch": 22.670025188916878,
+      "grad_norm": 5.8142499923706055,
+      "learning_rate": 1.1200807265388496e-05,
+      "loss": 6.0751,
+      "step": 54000
+    },
+    {
+      "epoch": 22.879932829554996,
+      "grad_norm": 3.877366781234741,
+      "learning_rate": 1.0191725529767912e-05,
+      "loss": 6.0692,
+      "step": 54500
+    },
+    {
+      "epoch": 23.089840470193113,
+      "grad_norm": 2.797621250152588,
+      "learning_rate": 9.182643794147326e-06,
+      "loss": 6.0592,
+      "step": 55000
+    },
+    {
+      "epoch": 23.299748110831235,
+      "grad_norm": 5.02012300491333,
+      "learning_rate": 8.173562058526741e-06,
+      "loss": 6.0509,
+      "step": 55500
+    },
+    {
+      "epoch": 23.509655751469353,
+      "grad_norm": 6.04094123840332,
+      "learning_rate": 7.164480322906156e-06,
+      "loss": 6.0568,
+      "step": 56000
+    },
+    {
+      "epoch": 23.719563392107474,
+      "grad_norm": 5.789729118347168,
+      "learning_rate": 6.155398587285571e-06,
+      "loss": 6.0485,
+      "step": 56500
+    },
+    {
+      "epoch": 23.929471032745592,
+      "grad_norm": 5.207681179046631,
+      "learning_rate": 5.146316851664986e-06,
+      "loss": 6.0544,
+      "step": 57000
+    },
+    {
+      "epoch": 24.13937867338371,
+      "grad_norm": 5.431936264038086,
+      "learning_rate": 4.1372351160444e-06,
+      "loss": 6.0438,
+      "step": 57500
+    },
+    {
+      "epoch": 24.349286314021832,
+      "grad_norm": 3.283940076828003,
+      "learning_rate": 3.1281533804238145e-06,
+      "loss": 6.0491,
+      "step": 58000
+    },
+    {
+      "epoch": 24.55919395465995,
+      "grad_norm": 5.0561299324035645,
+      "learning_rate": 2.1190716448032293e-06,
+      "loss": 6.0432,
+      "step": 58500
+    },
+    {
+      "epoch": 24.769101595298068,
+      "grad_norm": 4.42847204208374,
+      "learning_rate": 1.109989909182644e-06,
+      "loss": 6.0379,
+      "step": 59000
+    },
+    {
+      "epoch": 24.97900923593619,
+      "grad_norm": 3.86914324760437,
+      "learning_rate": 1.0090817356205853e-07,
+      "loss": 6.034,
+      "step": 59500
+    },
+    {
+      "epoch": 25.0,
+      "step": 59550,
+      "total_flos": 7.68154255975296e+17,
+      "train_loss": 6.938646208973716,
+      "train_runtime": 16943.8632,
+      "train_samples_per_second": 337.348,
+      "train_steps_per_second": 3.515
+    }
+  ],
+  "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": 7.68154255975296e+17,
+  "train_batch_size": 96,
+  "trial_name": null,
+  "trial_params": null
+}