Model save

README.md

@@ -1,11 +1,9 @@
 ---
 base_model: Qwen/Qwen2.5-1.5B-Instruct
-datasets: HuggingFaceH4/Bespoke-Stratos-17k
 library_name: transformers
 model_name: Qwen2.5-1.5B-Open-R1-Distill
 tags:
 - generated_from_trainer
-- open-r1
 - trl
 - sft
 licence: license
@@ -13,7 +11,7 @@ licence: license
 
 # Model Card for Qwen2.5-1.5B-Open-R1-Distill
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the [HuggingFaceH4/Bespoke-Stratos-17k](https://huggingface.co/datasets/HuggingFaceH4/Bespoke-Stratos-17k) dataset.
+This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -29,7 +27,7 @@ print(output["generated_text"])
 
 ## Training procedure
 
-[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/liangzizhangluo-hohai-university/huggingface/runs/4cxrwklk) 
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/liangzizhangluo-hohai-university/huggingface/runs/1vvgpyeu) 
 
 
 This model was trained with SFT.

all_results.json

@@ -1,8 +1,8 @@
 {
-    "total_flos": 28487712768000.0,
-    "train_loss": 0.920774197101593,
-    "train_runtime": 1759.2244,
-    "train_samples": 16610,
-    "train_samples_per_second": 18.19,
-    "train_steps_per_second": 0.568
+    "total_flos": 28457722773504.0,
+    "train_loss": 0.008928275344893336,
+    "train_runtime": 1701.2794,
+    "train_samples": 40,
+    "train_samples_per_second": 18.809,
+    "train_steps_per_second": 0.588
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "total_flos": 28487712768000.0,
-    "train_loss": 0.920774197101593,
-    "train_runtime": 1759.2244,
-    "train_samples": 16610,
-    "train_samples_per_second": 18.19,
-    "train_steps_per_second": 0.568
+    "total_flos": 28457722773504.0,
+    "train_loss": 0.008928275344893336,
+    "train_runtime": 1701.2794,
+    "train_samples": 40,
+    "train_samples_per_second": 18.809,
+    "train_steps_per_second": 0.588
 }

trainer_state.json

@@ -1,7 +1,7 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.3699593044765076,
+  "epoch": 500.0,
   "eval_steps": 500,
   "global_step": 1000,
   "is_hyper_param_search": false,
@@ -9,1419 +9,1419 @@
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.001849796522382538,
-      "grad_norm": 3.3279617254334535,
+      "epoch": 2.5,
+      "grad_norm": 8.514540942565027,
       "learning_rate": 5e-06,
-      "loss": 1.3195,
+      "loss": 0.585,
       "step": 5
     },
     {
-      "epoch": 0.003699593044765076,
-      "grad_norm": 1.9430817685675177,
+      "epoch": 5.0,
+      "grad_norm": 3.9267725944427414,
       "learning_rate": 1e-05,
-      "loss": 1.2699,
+      "loss": 0.4444,
       "step": 10
     },
     {
-      "epoch": 0.005549389567147614,
-      "grad_norm": 2.1226078391974967,
+      "epoch": 7.5,
+      "grad_norm": 1.359630389285586,
       "learning_rate": 1.5e-05,
-      "loss": 1.1007,
+      "loss": 0.1918,
       "step": 15
     },
     {
-      "epoch": 0.007399186089530152,
-      "grad_norm": 1.864202847273668,
+      "epoch": 10.0,
+      "grad_norm": 0.8675984069699776,
       "learning_rate": 2e-05,
-      "loss": 1.0631,
+      "loss": 0.0255,
       "step": 20
     },
     {
-      "epoch": 0.00924898261191269,
-      "grad_norm": 1.5790964667081508,
+      "epoch": 12.5,
+      "grad_norm": 1.3166063968433408,
       "learning_rate": 2.5e-05,
-      "loss": 1.0944,
+      "loss": 0.0094,
       "step": 25
     },
     {
-      "epoch": 0.011098779134295227,
-      "grad_norm": 1.3616177321523328,
+      "epoch": 15.0,
+      "grad_norm": 1.4298321029000574,
       "learning_rate": 3e-05,
-      "loss": 0.9914,
+      "loss": 0.0066,
       "step": 30
     },
     {
-      "epoch": 0.012948575656677765,
-      "grad_norm": 1.404161383781241,
+      "epoch": 17.5,
+      "grad_norm": 0.39016346214571596,
       "learning_rate": 3.5e-05,
-      "loss": 0.9922,
+      "loss": 0.0054,
       "step": 35
     },
     {
-      "epoch": 0.014798372179060304,
-      "grad_norm": 1.4817289036292525,
+      "epoch": 20.0,
+      "grad_norm": 0.2790201303316679,
       "learning_rate": 4e-05,
-      "loss": 0.992,
+      "loss": 0.0062,
       "step": 40
     },
     {
-      "epoch": 0.016648168701442843,
-      "grad_norm": 1.4532039393449283,
+      "epoch": 22.5,
+      "grad_norm": 0.08501619599523778,
       "learning_rate": 4.5e-05,
-      "loss": 0.9724,
+      "loss": 0.004,
       "step": 45
     },
     {
-      "epoch": 0.01849796522382538,
-      "grad_norm": 1.459454732453025,
+      "epoch": 25.0,
+      "grad_norm": 0.062067934755517584,
       "learning_rate": 5e-05,
-      "loss": 0.9584,
+      "loss": 0.0029,
       "step": 50
     },
     {
-      "epoch": 0.020347761746207917,
-      "grad_norm": 1.3435610444029484,
+      "epoch": 27.5,
+      "grad_norm": 0.052666305930186126,
       "learning_rate": 4.9996924362330555e-05,
-      "loss": 0.9809,
+      "loss": 0.0028,
       "step": 55
     },
     {
-      "epoch": 0.022197558268590455,
-      "grad_norm": 1.507556200878351,
+      "epoch": 30.0,
+      "grad_norm": 0.03938136111188392,
       "learning_rate": 4.998769829017084e-05,
-      "loss": 1.0205,
+      "loss": 0.0028,
       "step": 60
     },
     {
-      "epoch": 0.024047354790972992,
-      "grad_norm": 1.74928123764404,
+      "epoch": 32.5,
+      "grad_norm": 0.03394672644933696,
       "learning_rate": 4.997232430583686e-05,
-      "loss": 1.0078,
+      "loss": 0.0028,
       "step": 65
     },
     {
-      "epoch": 0.02589715131335553,
-      "grad_norm": 1.3343301345399712,
+      "epoch": 35.0,
+      "grad_norm": 0.037661243334742,
       "learning_rate": 4.995080661242243e-05,
-      "loss": 0.9691,
+      "loss": 0.0028,
       "step": 70
     },
     {
-      "epoch": 0.02774694783573807,
-      "grad_norm": 1.383358561734251,
+      "epoch": 37.5,
+      "grad_norm": 0.039184906370568255,
       "learning_rate": 4.992315109265007e-05,
-      "loss": 1.0215,
+      "loss": 0.0027,
       "step": 75
     },
     {
-      "epoch": 0.029596744358120607,
-      "grad_norm": 1.220791806067286,
+      "epoch": 40.0,
+      "grad_norm": 0.04435589843124744,
       "learning_rate": 4.988936530726276e-05,
-      "loss": 0.9797,
+      "loss": 0.0028,
       "step": 80
     },
     {
-      "epoch": 0.031446540880503145,
-      "grad_norm": 1.7288941628239176,
+      "epoch": 42.5,
+      "grad_norm": 0.02657151726999846,
       "learning_rate": 4.984945849295686e-05,
-      "loss": 0.9667,
+      "loss": 0.0026,
       "step": 85
     },
     {
-      "epoch": 0.033296337402885685,
-      "grad_norm": 1.3374822253676133,
+      "epoch": 45.0,
+      "grad_norm": 0.03472385835097059,
       "learning_rate": 4.980344155985695e-05,
-      "loss": 0.9725,
+      "loss": 0.0027,
       "step": 90
     },
     {
-      "epoch": 0.03514613392526822,
-      "grad_norm": 1.258960837768452,
+      "epoch": 47.5,
+      "grad_norm": 0.031070975559645563,
       "learning_rate": 4.975132708853304e-05,
-      "loss": 0.9877,
+      "loss": 0.0026,
       "step": 95
     },
     {
-      "epoch": 0.03699593044765076,
-      "grad_norm": 1.438691891865278,
+      "epoch": 50.0,
+      "grad_norm": 0.03730805984645713,
       "learning_rate": 4.9693129326561254e-05,
-      "loss": 1.0251,
+      "loss": 0.0027,
       "step": 100
     },
     {
-      "epoch": 0.038845726970033294,
-      "grad_norm": 1.3123491113835215,
+      "epoch": 52.5,
+      "grad_norm": 0.030401417138752773,
       "learning_rate": 4.96288641846286e-05,
-      "loss": 0.9805,
+      "loss": 0.0027,
       "step": 105
     },
     {
-      "epoch": 0.040695523492415835,
-      "grad_norm": 1.3868311724825364,
+      "epoch": 55.0,
+      "grad_norm": 0.030056720425659458,
       "learning_rate": 4.955854923218321e-05,
-      "loss": 0.9866,
+      "loss": 0.0027,
       "step": 110
     },
     {
-      "epoch": 0.042545320014798375,
-      "grad_norm": 1.1867492456163233,
+      "epoch": 57.5,
+      "grad_norm": 0.02714242610959012,
       "learning_rate": 4.948220369263099e-05,
-      "loss": 0.9671,
+      "loss": 0.0025,
       "step": 115
     },
     {
-      "epoch": 0.04439511653718091,
-      "grad_norm": 1.3516961645638472,
+      "epoch": 60.0,
+      "grad_norm": 0.021468304046160087,
       "learning_rate": 4.939984843808013e-05,
-      "loss": 0.9712,
+      "loss": 0.0026,
       "step": 120
     },
     {
-      "epoch": 0.04624491305956345,
-      "grad_norm": 1.3818877123258775,
+      "epoch": 62.5,
+      "grad_norm": 0.027579476441235425,
       "learning_rate": 4.931150598363494e-05,
-      "loss": 0.9973,
+      "loss": 0.0027,
       "step": 125
     },
     {
-      "epoch": 0.048094709581945984,
-      "grad_norm": 1.2461243722970794,
+      "epoch": 65.0,
+      "grad_norm": 0.041471384473431336,
       "learning_rate": 4.921720048124035e-05,
-      "loss": 0.9721,
+      "loss": 0.0028,
       "step": 130
     },
     {
-      "epoch": 0.049944506104328525,
-      "grad_norm": 1.3531497904983025,
+      "epoch": 67.5,
+      "grad_norm": 0.030743428074535287,
       "learning_rate": 4.9116957713079084e-05,
-      "loss": 0.9947,
+      "loss": 0.0027,
       "step": 135
     },
     {
-      "epoch": 0.05179430262671106,
-      "grad_norm": 1.3909365555868392,
+      "epoch": 70.0,
+      "grad_norm": 0.03200739761669791,
       "learning_rate": 4.901080508452305e-05,
-      "loss": 0.9886,
+      "loss": 0.0027,
       "step": 140
     },
     {
-      "epoch": 0.0536440991490936,
-      "grad_norm": 1.3632639733726144,
+      "epoch": 72.5,
+      "grad_norm": 0.02266656589663142,
       "learning_rate": 4.889877161664096e-05,
-      "loss": 0.9856,
+      "loss": 0.0027,
       "step": 145
     },
     {
-      "epoch": 0.05549389567147614,
-      "grad_norm": 1.376746573930021,
+      "epoch": 75.0,
+      "grad_norm": 0.03858968870626217,
       "learning_rate": 4.878088793826428e-05,
-      "loss": 1.0124,
+      "loss": 0.0026,
       "step": 150
     },
     {
-      "epoch": 0.057343692193858674,
-      "grad_norm": 1.2051488416511116,
+      "epoch": 77.5,
+      "grad_norm": 0.0436229252420695,
       "learning_rate": 4.865718627761363e-05,
-      "loss": 0.9502,
+      "loss": 0.0027,
       "step": 155
     },
     {
-      "epoch": 0.059193488716241215,
-      "grad_norm": 1.4444748706679968,
+      "epoch": 80.0,
+      "grad_norm": 0.021505547895787954,
       "learning_rate": 4.8527700453487873e-05,
-      "loss": 0.976,
+      "loss": 0.0026,
       "step": 160
     },
     {
-      "epoch": 0.06104328523862375,
-      "grad_norm": 1.5080951955787747,
+      "epoch": 82.5,
+      "grad_norm": 0.020985544413640905,
       "learning_rate": 4.839246586601841e-05,
-      "loss": 0.9858,
+      "loss": 0.0027,
       "step": 165
     },
     {
-      "epoch": 0.06289308176100629,
-      "grad_norm": 1.218432538840472,
+      "epoch": 85.0,
+      "grad_norm": 0.023894609006279273,
       "learning_rate": 4.825151948699116e-05,
-      "loss": 0.985,
+      "loss": 0.0028,
       "step": 170
     },
     {
-      "epoch": 0.06474287828338883,
-      "grad_norm": 1.1799037097271943,
+      "epoch": 87.5,
+      "grad_norm": 0.02603189523521318,
       "learning_rate": 4.8104899849738795e-05,
-      "loss": 0.9378,
+      "loss": 0.0026,
       "step": 175
     },
     {
-      "epoch": 0.06659267480577137,
-      "grad_norm": 1.2779121353402405,
+      "epoch": 90.0,
+      "grad_norm": 0.033072587670050196,
       "learning_rate": 4.795264703860616e-05,
-      "loss": 0.9683,
+      "loss": 0.0027,
       "step": 180
     },
     {
-      "epoch": 0.0684424713281539,
-      "grad_norm": 1.4722942379840678,
+      "epoch": 92.5,
+      "grad_norm": 0.029455707311945873,
       "learning_rate": 4.779480267799158e-05,
-      "loss": 0.9904,
+      "loss": 0.0027,
       "step": 185
     },
     {
-      "epoch": 0.07029226785053644,
-      "grad_norm": 1.3984538759925134,
+      "epoch": 95.0,
+      "grad_norm": 0.031066744462249135,
       "learning_rate": 4.763140992096718e-05,
-      "loss": 0.9702,
+      "loss": 0.0027,
       "step": 190
     },
     {
-      "epoch": 0.07214206437291898,
-      "grad_norm": 1.2519064402447997,
+      "epoch": 97.5,
+      "grad_norm": 0.03179467575788789,
       "learning_rate": 4.7462513437481266e-05,
-      "loss": 0.9671,
+      "loss": 0.0027,
       "step": 195
     },
     {
-      "epoch": 0.07399186089530152,
-      "grad_norm": 1.281644209465442,
+      "epoch": 100.0,
+      "grad_norm": 0.04391757622955937,
       "learning_rate": 4.7288159402146e-05,
-      "loss": 0.9653,
+      "loss": 0.0027,
       "step": 200
     },
     {
-      "epoch": 0.07584165741768406,
-      "grad_norm": 1.3861942696753027,
+      "epoch": 102.5,
+      "grad_norm": 0.029394885252518814,
       "learning_rate": 4.7108395481613736e-05,
-      "loss": 0.989,
+      "loss": 0.0028,
       "step": 205
     },
     {
-      "epoch": 0.07769145394006659,
-      "grad_norm": 1.2692148251690907,
+      "epoch": 105.0,
+      "grad_norm": 0.027320800240786685,
       "learning_rate": 4.692327082154542e-05,
-      "loss": 0.9695,
+      "loss": 0.0027,
       "step": 210
     },
     {
-      "epoch": 0.07954125046244913,
-      "grad_norm": 1.2503784332879924,
+      "epoch": 107.5,
+      "grad_norm": 0.030873484414188097,
       "learning_rate": 4.6732836033174634e-05,
-      "loss": 0.9442,
+      "loss": 0.0027,
       "step": 215
     },
     {
-      "epoch": 0.08139104698483167,
-      "grad_norm": 1.2071851181770694,
+      "epoch": 110.0,
+      "grad_norm": 0.048697733723929194,
       "learning_rate": 4.653714317947105e-05,
-      "loss": 0.9664,
+      "loss": 0.0027,
       "step": 220
     },
     {
-      "epoch": 0.08324084350721421,
-      "grad_norm": 1.3973898482595015,
+      "epoch": 112.5,
+      "grad_norm": 0.025217762723189756,
       "learning_rate": 4.6336245760906896e-05,
-      "loss": 1.0037,
+      "loss": 0.0026,
       "step": 225
     },
     {
-      "epoch": 0.08509064002959675,
-      "grad_norm": 1.1885748014842101,
+      "epoch": 115.0,
+      "grad_norm": 0.034325242650264584,
       "learning_rate": 4.6130198700830455e-05,
-      "loss": 0.9523,
+      "loss": 0.0027,
       "step": 230
     },
     {
-      "epoch": 0.08694043655197928,
-      "grad_norm": 1.1676588738483402,
+      "epoch": 117.5,
+      "grad_norm": 0.0392844047749773,
       "learning_rate": 4.591905833045059e-05,
-      "loss": 0.9317,
+      "loss": 0.0027,
       "step": 235
     },
     {
-      "epoch": 0.08879023307436182,
-      "grad_norm": 1.2482898560322035,
+      "epoch": 120.0,
+      "grad_norm": 0.02965200107252582,
       "learning_rate": 4.570288237343632e-05,
-      "loss": 0.9315,
+      "loss": 0.0026,
       "step": 240
     },
     {
-      "epoch": 0.09064002959674436,
-      "grad_norm": 1.122744743080674,
+      "epoch": 122.5,
+      "grad_norm": 0.020210067730727373,
       "learning_rate": 4.5481729930135805e-05,
-      "loss": 0.9582,
+      "loss": 0.0026,
       "step": 245
     },
     {
-      "epoch": 0.0924898261191269,
-      "grad_norm": 1.1964791201941039,
+      "epoch": 125.0,
+      "grad_norm": 0.027623657423726022,
       "learning_rate": 4.5255661461418854e-05,
-      "loss": 0.9613,
+      "loss": 0.0026,
       "step": 250
     },
     {
-      "epoch": 0.09433962264150944,
-      "grad_norm": 1.313613853504735,
+      "epoch": 127.5,
+      "grad_norm": 0.023255879069235798,
       "learning_rate": 4.502473877214754e-05,
-      "loss": 0.9627,
+      "loss": 0.0027,
       "step": 255
     },
     {
-      "epoch": 0.09618941916389197,
-      "grad_norm": 1.2998325386052085,
+      "epoch": 130.0,
+      "grad_norm": 0.014768241964023569,
       "learning_rate": 4.478902499427933e-05,
-      "loss": 0.9435,
+      "loss": 0.0027,
       "step": 260
     },
     {
-      "epoch": 0.09803921568627451,
-      "grad_norm": 1.1188167692403637,
+      "epoch": 132.5,
+      "grad_norm": 0.03668790355624439,
       "learning_rate": 4.454858456960754e-05,
-      "loss": 0.9531,
+      "loss": 0.0027,
       "step": 265
     },
     {
-      "epoch": 0.09988901220865705,
-      "grad_norm": 1.245499023500759,
+      "epoch": 135.0,
+      "grad_norm": 0.028968111779672595,
       "learning_rate": 4.430348323214347e-05,
-      "loss": 0.9615,
+      "loss": 0.0026,
       "step": 270
     },
     {
-      "epoch": 0.10173880873103959,
-      "grad_norm": 1.2731826201754,
+      "epoch": 137.5,
+      "grad_norm": 0.02287695653834384,
       "learning_rate": 4.4053787990145465e-05,
-      "loss": 0.9852,
+      "loss": 0.0027,
       "step": 275
     },
     {
-      "epoch": 0.10358860525342212,
-      "grad_norm": 1.0779984323652068,
+      "epoch": 140.0,
+      "grad_norm": 0.014260556347040123,
       "learning_rate": 4.379956710779951e-05,
-      "loss": 0.9649,
+      "loss": 0.0027,
       "step": 280
     },
     {
-      "epoch": 0.10543840177580466,
-      "grad_norm": 1.3267240330921417,
+      "epoch": 142.5,
+      "grad_norm": 0.029771289364298407,
       "learning_rate": 4.3540890086556435e-05,
-      "loss": 0.967,
+      "loss": 0.0026,
       "step": 285
     },
     {
-      "epoch": 0.1072881982981872,
-      "grad_norm": 1.194869135121013,
+      "epoch": 145.0,
+      "grad_norm": 0.03788961751665883,
       "learning_rate": 4.327782764613099e-05,
-      "loss": 0.9745,
+      "loss": 0.0027,
       "step": 290
     },
     {
-      "epoch": 0.10913799482056974,
-      "grad_norm": 1.092064798384306,
+      "epoch": 147.5,
+      "grad_norm": 0.03971119187487509,
       "learning_rate": 4.301045170516773e-05,
-      "loss": 0.9354,
+      "loss": 0.0026,
       "step": 295
     },
     {
-      "epoch": 0.11098779134295228,
-      "grad_norm": 1.1564826438676266,
+      "epoch": 150.0,
+      "grad_norm": 0.03370808514177945,
       "learning_rate": 4.2738835361579175e-05,
-      "loss": 0.9301,
+      "loss": 0.0026,
       "step": 300
     },
     {
-      "epoch": 0.1128375878653348,
-      "grad_norm": 1.1034286954430845,
+      "epoch": 152.5,
+      "grad_norm": 0.04042652697228835,
       "learning_rate": 4.2463052872561584e-05,
-      "loss": 0.9536,
+      "loss": 0.0027,
       "step": 305
     },
     {
-      "epoch": 0.11468738438771735,
-      "grad_norm": 1.1101601059988588,
+      "epoch": 155.0,
+      "grad_norm": 0.03521656241677526,
       "learning_rate": 4.2183179634293794e-05,
-      "loss": 0.9302,
+      "loss": 0.0028,
       "step": 310
     },
     {
-      "epoch": 0.11653718091009989,
-      "grad_norm": 1.2757853825149674,
+      "epoch": 157.5,
+      "grad_norm": 0.03092781528226213,
       "learning_rate": 4.1899292161324627e-05,
-      "loss": 0.9547,
+      "loss": 0.0026,
       "step": 315
     },
     {
-      "epoch": 0.11838697743248243,
-      "grad_norm": 1.1087176065384894,
+      "epoch": 160.0,
+      "grad_norm": 0.020310997173564212,
       "learning_rate": 4.1611468065654586e-05,
-      "loss": 0.9764,
+      "loss": 0.0026,
       "step": 320
     },
     {
-      "epoch": 0.12023677395486497,
-      "grad_norm": 1.2270496050163748,
+      "epoch": 162.5,
+      "grad_norm": 0.015790567180341695,
       "learning_rate": 4.1319786035517534e-05,
-      "loss": 0.9551,
+      "loss": 0.0027,
       "step": 325
     },
     {
-      "epoch": 0.1220865704772475,
-      "grad_norm": 1.2025590067136398,
+      "epoch": 165.0,
+      "grad_norm": 0.0194147951616345,
       "learning_rate": 4.1024325813868065e-05,
-      "loss": 0.9238,
+      "loss": 0.0027,
       "step": 330
     },
     {
-      "epoch": 0.12393636699963004,
-      "grad_norm": 1.1631866711130052,
+      "epoch": 167.5,
+      "grad_norm": 0.030730482667734805,
       "learning_rate": 4.072516817658065e-05,
-      "loss": 0.9573,
+      "loss": 0.0027,
       "step": 335
     },
     {
-      "epoch": 0.12578616352201258,
-      "grad_norm": 1.064949988724441,
+      "epoch": 170.0,
+      "grad_norm": 0.029837731953442564,
       "learning_rate": 4.0422394910366236e-05,
-      "loss": 0.9184,
+      "loss": 0.0026,
       "step": 340
     },
     {
-      "epoch": 0.12763596004439512,
-      "grad_norm": 1.4077051494514279,
+      "epoch": 172.5,
+      "grad_norm": 0.02492166177013941,
       "learning_rate": 4.0116088790412645e-05,
-      "loss": 0.9518,
+      "loss": 0.0026,
       "step": 345
     },
     {
-      "epoch": 0.12948575656677766,
-      "grad_norm": 1.2486218937453641,
+      "epoch": 175.0,
+      "grad_norm": 0.022790129612325603,
       "learning_rate": 3.980633355775461e-05,
-      "loss": 0.9518,
+      "loss": 0.0027,
       "step": 350
     },
     {
-      "epoch": 0.1313355530891602,
-      "grad_norm": 1.2416341494672163,
+      "epoch": 177.5,
+      "grad_norm": 0.0367286949474086,
       "learning_rate": 3.949321389637986e-05,
-      "loss": 0.9238,
+      "loss": 0.0027,
       "step": 355
     },
     {
-      "epoch": 0.13318534961154274,
-      "grad_norm": 1.18527481990195,
+      "epoch": 180.0,
+      "grad_norm": 0.026449210914021956,
       "learning_rate": 3.917681541007734e-05,
-      "loss": 0.9319,
+      "loss": 0.0025,
       "step": 360
     },
     {
-      "epoch": 0.13503514613392525,
-      "grad_norm": 1.1856076287802888,
+      "epoch": 182.5,
+      "grad_norm": 0.025906355217603764,
       "learning_rate": 3.885722459903399e-05,
-      "loss": 0.941,
+      "loss": 0.0027,
       "step": 365
     },
     {
-      "epoch": 0.1368849426563078,
-      "grad_norm": 1.3096185432838516,
+      "epoch": 185.0,
+      "grad_norm": 0.013510506972149507,
       "learning_rate": 3.853452883618644e-05,
-      "loss": 0.9531,
+      "loss": 0.0027,
       "step": 370
     },
     {
-      "epoch": 0.13873473917869034,
-      "grad_norm": 1.0667136184374695,
+      "epoch": 187.5,
+      "grad_norm": 0.02126782273762181,
       "learning_rate": 3.8208816343334156e-05,
-      "loss": 0.9648,
+      "loss": 0.0027,
       "step": 375
     },
     {
-      "epoch": 0.14058453570107288,
-      "grad_norm": 1.3556343012265513,
+      "epoch": 190.0,
+      "grad_norm": 0.02778025887186039,
       "learning_rate": 3.788017616702048e-05,
-      "loss": 0.9622,
+      "loss": 0.0027,
       "step": 380
     },
     {
-      "epoch": 0.14243433222345542,
-      "grad_norm": 1.2826346418353032,
+      "epoch": 192.5,
+      "grad_norm": 0.02023158670094667,
       "learning_rate": 3.754869815418815e-05,
-      "loss": 0.9243,
+      "loss": 0.0026,
       "step": 385
     },
     {
-      "epoch": 0.14428412874583796,
-      "grad_norm": 1.4908193033444779,
+      "epoch": 195.0,
+      "grad_norm": 0.031240526748051056,
       "learning_rate": 3.721447292761609e-05,
-      "loss": 0.9402,
+      "loss": 0.0027,
       "step": 390
     },
     {
-      "epoch": 0.1461339252682205,
-      "grad_norm": 1.2092388640077134,
+      "epoch": 197.5,
+      "grad_norm": 0.02268049791605829,
       "learning_rate": 3.687759186114403e-05,
-      "loss": 0.8881,
+      "loss": 0.0026,
       "step": 395
     },
     {
-      "epoch": 0.14798372179060304,
-      "grad_norm": 1.2289784753028945,
+      "epoch": 200.0,
+      "grad_norm": 0.02998474683024178,
       "learning_rate": 3.6538147054691817e-05,
-      "loss": 0.9171,
+      "loss": 0.0027,
       "step": 400
     },
     {
-      "epoch": 0.14983351831298558,
-      "grad_norm": 1.2686814293232442,
+      "epoch": 202.5,
+      "grad_norm": 0.026430986104193692,
       "learning_rate": 3.619623130908018e-05,
-      "loss": 0.9487,
+      "loss": 0.0027,
       "step": 405
     },
     {
-      "epoch": 0.15168331483536812,
-      "grad_norm": 1.0701718673112672,
+      "epoch": 205.0,
+      "grad_norm": 0.010605276850508178,
       "learning_rate": 3.5851938100659964e-05,
-      "loss": 0.9099,
+      "loss": 0.0027,
       "step": 410
     },
     {
-      "epoch": 0.15353311135775063,
-      "grad_norm": 1.090800531105658,
+      "epoch": 207.5,
+      "grad_norm": 0.02068209169464807,
       "learning_rate": 3.550536155575662e-05,
-      "loss": 0.8853,
+      "loss": 0.0027,
       "step": 415
     },
     {
-      "epoch": 0.15538290788013318,
-      "grad_norm": 1.2313388220061225,
+      "epoch": 210.0,
+      "grad_norm": 0.03300339570363238,
       "learning_rate": 3.515659642493697e-05,
-      "loss": 0.9385,
+      "loss": 0.0026,
       "step": 420
     },
     {
-      "epoch": 0.15723270440251572,
-      "grad_norm": 1.2536700830777894,
+      "epoch": 212.5,
+      "grad_norm": 0.026016657903437497,
       "learning_rate": 3.480573805710538e-05,
-      "loss": 0.9143,
+      "loss": 0.0027,
       "step": 425
     },
     {
-      "epoch": 0.15908250092489826,
-      "grad_norm": 1.1514860580661626,
+      "epoch": 215.0,
+      "grad_norm": 0.024770792616580162,
       "learning_rate": 3.4452882373436316e-05,
-      "loss": 0.8996,
+      "loss": 0.0026,
       "step": 430
     },
     {
-      "epoch": 0.1609322974472808,
-      "grad_norm": 1.3731197382175357,
+      "epoch": 217.5,
+      "grad_norm": 0.0265444448484458,
       "learning_rate": 3.4098125841150466e-05,
-      "loss": 0.9243,
+      "loss": 0.0026,
       "step": 435
     },
     {
-      "epoch": 0.16278209396966334,
-      "grad_norm": 1.0232827661891992,
+      "epoch": 220.0,
+      "grad_norm": 0.01924921498743014,
       "learning_rate": 3.37415654471415e-05,
-      "loss": 0.9369,
+      "loss": 0.0026,
       "step": 440
     },
     {
-      "epoch": 0.16463189049204588,
-      "grad_norm": 1.1305694231647365,
+      "epoch": 222.5,
+      "grad_norm": 0.027603952267821173,
       "learning_rate": 3.3383298671460944e-05,
-      "loss": 0.9142,
+      "loss": 0.0026,
       "step": 445
     },
     {
-      "epoch": 0.16648168701442842,
-      "grad_norm": 1.1398463628627697,
+      "epoch": 225.0,
+      "grad_norm": 0.0254709928199966,
       "learning_rate": 3.3023423460667985e-05,
-      "loss": 0.9094,
+      "loss": 0.0026,
       "step": 450
     },
     {
-      "epoch": 0.16833148353681096,
-      "grad_norm": 1.0590670030120253,
+      "epoch": 227.5,
+      "grad_norm": 0.031183288661866662,
       "learning_rate": 3.2662038201051914e-05,
-      "loss": 0.9013,
+      "loss": 0.0027,
       "step": 455
     },
     {
-      "epoch": 0.1701812800591935,
-      "grad_norm": 1.1536632313067792,
+      "epoch": 230.0,
+      "grad_norm": 0.03425285898227886,
       "learning_rate": 3.2299241691734304e-05,
-      "loss": 0.9233,
+      "loss": 0.0026,
       "step": 460
     },
     {
-      "epoch": 0.17203107658157601,
-      "grad_norm": 1.167439193543511,
+      "epoch": 232.5,
+      "grad_norm": 0.028752151160391863,
       "learning_rate": 3.1935133117658306e-05,
-      "loss": 0.952,
+      "loss": 0.0027,
       "step": 465
     },
     {
-      "epoch": 0.17388087310395856,
-      "grad_norm": 1.1956939577700672,
+      "epoch": 235.0,
+      "grad_norm": 0.028976330470022123,
       "learning_rate": 3.156981202247248e-05,
-      "loss": 0.9247,
+      "loss": 0.0026,
       "step": 470
     },
     {
-      "epoch": 0.1757306696263411,
-      "grad_norm": 1.1593001767390334,
+      "epoch": 237.5,
+      "grad_norm": 0.026920539664030775,
       "learning_rate": 3.1203378281316515e-05,
-      "loss": 0.9012,
+      "loss": 0.0026,
       "step": 475
     },
     {
-      "epoch": 0.17758046614872364,
-      "grad_norm": 1.0435734696221646,
+      "epoch": 240.0,
+      "grad_norm": 0.020989002709489214,
       "learning_rate": 3.0835932073516444e-05,
-      "loss": 0.9154,
+      "loss": 0.0027,
       "step": 480
     },
     {
-      "epoch": 0.17943026267110618,
-      "grad_norm": 1.1108156220083871,
+      "epoch": 242.5,
+      "grad_norm": 0.024523095658861207,
       "learning_rate": 3.0467573855196558e-05,
-      "loss": 0.899,
+      "loss": 0.0027,
       "step": 485
     },
     {
-      "epoch": 0.18128005919348872,
-      "grad_norm": 1.0167505348335453,
+      "epoch": 245.0,
+      "grad_norm": 0.015580401724056997,
       "learning_rate": 3.0098404331815695e-05,
-      "loss": 0.9174,
+      "loss": 0.0027,
       "step": 490
     },
     {
-      "epoch": 0.18312985571587126,
-      "grad_norm": 0.965632547470374,
+      "epoch": 247.5,
+      "grad_norm": 0.022996840296875175,
       "learning_rate": 2.9728524430635417e-05,
-      "loss": 0.9143,
+      "loss": 0.0027,
       "step": 495
     },
     {
-      "epoch": 0.1849796522382538,
-      "grad_norm": 1.1035315117198525,
+      "epoch": 250.0,
+      "grad_norm": 0.021854877708908236,
       "learning_rate": 2.9358035273127483e-05,
-      "loss": 0.9261,
+      "loss": 0.0026,
       "step": 500
     },
     {
-      "epoch": 0.18682944876063634,
-      "grad_norm": 1.1169822035484065,
+      "epoch": 252.5,
+      "grad_norm": 0.025003129400125675,
       "learning_rate": 2.8987038147328238e-05,
-      "loss": 0.9313,
+      "loss": 0.0026,
       "step": 505
     },
     {
-      "epoch": 0.18867924528301888,
-      "grad_norm": 0.9769048796586226,
+      "epoch": 255.0,
+      "grad_norm": 0.02759119149124416,
       "learning_rate": 2.86156344801475e-05,
-      "loss": 0.9132,
+      "loss": 0.0026,
       "step": 510
     },
     {
-      "epoch": 0.1905290418054014,
-      "grad_norm": 1.030443130117634,
+      "epoch": 257.5,
+      "grad_norm": 0.024349103530368846,
       "learning_rate": 2.824392580963944e-05,
-      "loss": 0.9115,
+      "loss": 0.0027,
       "step": 515
     },
     {
-      "epoch": 0.19237883832778394,
-      "grad_norm": 1.1062439088724974,
+      "epoch": 260.0,
+      "grad_norm": 0.02918200596645593,
       "learning_rate": 2.787201375724307e-05,
-      "loss": 0.884,
+      "loss": 0.0027,
       "step": 520
     },
     {
-      "epoch": 0.19422863485016648,
-      "grad_norm": 1.0597749922719453,
+      "epoch": 262.5,
+      "grad_norm": 0.017661985537795705,
       "learning_rate": 2.7500000000000004e-05,
-      "loss": 0.8783,
+      "loss": 0.0026,
       "step": 525
     },
     {
-      "epoch": 0.19607843137254902,
-      "grad_norm": 1.1259858455874245,
+      "epoch": 265.0,
+      "grad_norm": 0.02880881596677379,
       "learning_rate": 2.7127986242756936e-05,
-      "loss": 0.8982,
+      "loss": 0.0026,
       "step": 530
     },
     {
-      "epoch": 0.19792822789493156,
-      "grad_norm": 1.1500074530646092,
+      "epoch": 267.5,
+      "grad_norm": 0.04090166133876468,
       "learning_rate": 2.6756074190360563e-05,
-      "loss": 0.9127,
+      "loss": 0.0028,
       "step": 535
     },
     {
-      "epoch": 0.1997780244173141,
-      "grad_norm": 1.0395549080430184,
+      "epoch": 270.0,
+      "grad_norm": 0.022838636369005745,
       "learning_rate": 2.63843655198525e-05,
-      "loss": 0.8975,
+      "loss": 0.0027,
       "step": 540
     },
     {
-      "epoch": 0.20162782093969664,
-      "grad_norm": 1.083025795794125,
+      "epoch": 272.5,
+      "grad_norm": 0.03417808915995595,
       "learning_rate": 2.6012961852671767e-05,
-      "loss": 0.9231,
+      "loss": 0.0026,
       "step": 545
     },
     {
-      "epoch": 0.20347761746207918,
-      "grad_norm": 1.0839550256360933,
+      "epoch": 275.0,
+      "grad_norm": 0.02391514801170413,
       "learning_rate": 2.5641964726872526e-05,
-      "loss": 0.9459,
+      "loss": 0.0026,
       "step": 550
     },
     {
-      "epoch": 0.20532741398446172,
-      "grad_norm": 1.0874256330778362,
+      "epoch": 277.5,
+      "grad_norm": 0.027050966559574396,
       "learning_rate": 2.527147556936459e-05,
-      "loss": 0.891,
+      "loss": 0.0027,
       "step": 555
     },
     {
-      "epoch": 0.20717721050684423,
-      "grad_norm": 1.19750486880985,
+      "epoch": 280.0,
+      "grad_norm": 0.029392338368449532,
       "learning_rate": 2.4901595668184314e-05,
-      "loss": 0.8838,
+      "loss": 0.0027,
       "step": 560
     },
     {
-      "epoch": 0.20902700702922677,
-      "grad_norm": 1.092576941889689,
+      "epoch": 282.5,
+      "grad_norm": 0.03213320609449678,
       "learning_rate": 2.453242614480345e-05,
-      "loss": 0.9442,
+      "loss": 0.0027,
       "step": 565
     },
     {
-      "epoch": 0.21087680355160932,
-      "grad_norm": 1.227755705501534,
+      "epoch": 285.0,
+      "grad_norm": 0.017580210273689653,
       "learning_rate": 2.416406792648355e-05,
-      "loss": 0.8934,
+      "loss": 0.0027,
       "step": 570
     },
     {
-      "epoch": 0.21272660007399186,
-      "grad_norm": 1.2087319750759349,
+      "epoch": 287.5,
+      "grad_norm": 0.025395175371267904,
       "learning_rate": 2.3796621718683487e-05,
-      "loss": 0.9035,
+      "loss": 0.0027,
       "step": 575
     },
     {
-      "epoch": 0.2145763965963744,
-      "grad_norm": 1.1471923615461788,
+      "epoch": 290.0,
+      "grad_norm": 0.02830946439085919,
       "learning_rate": 2.3430187977527533e-05,
-      "loss": 0.8942,
+      "loss": 0.0027,
       "step": 580
     },
     {
-      "epoch": 0.21642619311875694,
-      "grad_norm": 0.9623686085619546,
+      "epoch": 292.5,
+      "grad_norm": 0.04065072947108163,
       "learning_rate": 2.3064866882341696e-05,
-      "loss": 0.8907,
+      "loss": 0.0027,
       "step": 585
     },
     {
-      "epoch": 0.21827598964113948,
-      "grad_norm": 1.1021625170926432,
+      "epoch": 295.0,
+      "grad_norm": 0.028524952343665373,
       "learning_rate": 2.27007583082657e-05,
-      "loss": 0.8488,
+      "loss": 0.0026,
       "step": 590
     },
     {
-      "epoch": 0.22012578616352202,
-      "grad_norm": 1.0507739548200454,
+      "epoch": 297.5,
+      "grad_norm": 0.0284550999234347,
       "learning_rate": 2.233796179894809e-05,
-      "loss": 0.9243,
+      "loss": 0.0026,
       "step": 595
     },
     {
-      "epoch": 0.22197558268590456,
-      "grad_norm": 1.0241361415980714,
+      "epoch": 300.0,
+      "grad_norm": 0.03018543474602915,
       "learning_rate": 2.1976576539332024e-05,
-      "loss": 0.8746,
+      "loss": 0.0026,
       "step": 600
     },
     {
-      "epoch": 0.2238253792082871,
-      "grad_norm": 1.1308505070134895,
+      "epoch": 302.5,
+      "grad_norm": 0.017475284356752212,
       "learning_rate": 2.1616701328539057e-05,
-      "loss": 0.9094,
+      "loss": 0.0026,
       "step": 605
     },
     {
-      "epoch": 0.2256751757306696,
-      "grad_norm": 1.0945099014034436,
+      "epoch": 305.0,
+      "grad_norm": 0.02381743267091071,
       "learning_rate": 2.1258434552858502e-05,
-      "loss": 0.9125,
+      "loss": 0.0026,
       "step": 610
     },
     {
-      "epoch": 0.22752497225305215,
-      "grad_norm": 1.116483983505738,
+      "epoch": 307.5,
+      "grad_norm": 0.030282727334382083,
       "learning_rate": 2.090187415884955e-05,
-      "loss": 0.9148,
+      "loss": 0.0027,
       "step": 615
     },
     {
-      "epoch": 0.2293747687754347,
-      "grad_norm": 0.9728451267819282,
+      "epoch": 310.0,
+      "grad_norm": 0.02239088576419377,
       "learning_rate": 2.054711762656369e-05,
-      "loss": 0.8596,
+      "loss": 0.0026,
       "step": 620
     },
     {
-      "epoch": 0.23122456529781724,
-      "grad_norm": 1.0747467852239778,
+      "epoch": 312.5,
+      "grad_norm": 0.020772310557568244,
       "learning_rate": 2.0194261942894628e-05,
-      "loss": 0.8733,
+      "loss": 0.0026,
       "step": 625
     },
     {
-      "epoch": 0.23307436182019978,
-      "grad_norm": 1.0108414553651262,
+      "epoch": 315.0,
+      "grad_norm": 0.03490126977201953,
       "learning_rate": 1.984340357506303e-05,
-      "loss": 0.8476,
+      "loss": 0.0026,
       "step": 630
     },
     {
-      "epoch": 0.23492415834258232,
-      "grad_norm": 1.0393062726091873,
+      "epoch": 317.5,
+      "grad_norm": 0.015968447976387346,
       "learning_rate": 1.949463844424338e-05,
-      "loss": 0.8926,
+      "loss": 0.0026,
       "step": 635
     },
     {
-      "epoch": 0.23677395486496486,
-      "grad_norm": 1.017203795362952,
+      "epoch": 320.0,
+      "grad_norm": 0.023986554939275363,
       "learning_rate": 1.914806189934003e-05,
-      "loss": 0.8905,
+      "loss": 0.0028,
       "step": 640
     },
     {
-      "epoch": 0.2386237513873474,
-      "grad_norm": 1.0799681224310813,
+      "epoch": 322.5,
+      "grad_norm": 0.03420061182554338,
       "learning_rate": 1.8803768690919832e-05,
-      "loss": 0.8897,
+      "loss": 0.0027,
       "step": 645
     },
     {
-      "epoch": 0.24047354790972994,
-      "grad_norm": 1.0781213573400381,
+      "epoch": 325.0,
+      "grad_norm": 0.01792951575408546,
       "learning_rate": 1.8461852945308196e-05,
-      "loss": 0.8637,
+      "loss": 0.0027,
       "step": 650
     },
     {
-      "epoch": 0.24232334443211248,
-      "grad_norm": 1.1249251204928792,
+      "epoch": 327.5,
+      "grad_norm": 0.023802622532695907,
       "learning_rate": 1.8122408138855974e-05,
-      "loss": 0.8802,
+      "loss": 0.0026,
       "step": 655
     },
     {
-      "epoch": 0.244173140954495,
-      "grad_norm": 1.1923220257639287,
+      "epoch": 330.0,
+      "grad_norm": 0.023316524889500846,
       "learning_rate": 1.778552707238391e-05,
-      "loss": 0.9033,
+      "loss": 0.0026,
       "step": 660
     },
     {
-      "epoch": 0.24602293747687753,
-      "grad_norm": 1.0225563377110887,
+      "epoch": 332.5,
+      "grad_norm": 0.013376585581553524,
       "learning_rate": 1.7451301845811857e-05,
-      "loss": 0.8605,
+      "loss": 0.0026,
       "step": 665
     },
     {
-      "epoch": 0.24787273399926008,
-      "grad_norm": 1.0757446768121959,
+      "epoch": 335.0,
+      "grad_norm": 0.019080334175566694,
       "learning_rate": 1.711982383297953e-05,
-      "loss": 0.8906,
+      "loss": 0.0026,
       "step": 670
     },
     {
-      "epoch": 0.24972253052164262,
-      "grad_norm": 1.0589738260259443,
+      "epoch": 337.5,
+      "grad_norm": 0.01654436855407368,
       "learning_rate": 1.6791183656665846e-05,
-      "loss": 0.8938,
+      "loss": 0.0026,
       "step": 675
     },
     {
-      "epoch": 0.25157232704402516,
-      "grad_norm": 0.9884710814924109,
+      "epoch": 340.0,
+      "grad_norm": 0.022278657387787754,
       "learning_rate": 1.6465471163813574e-05,
-      "loss": 0.9306,
+      "loss": 0.0027,
       "step": 680
     },
     {
-      "epoch": 0.2534221235664077,
-      "grad_norm": 1.0526270648654439,
+      "epoch": 342.5,
+      "grad_norm": 0.015375948509190046,
       "learning_rate": 1.6142775400966023e-05,
-      "loss": 0.9038,
+      "loss": 0.0027,
       "step": 685
     },
     {
-      "epoch": 0.25527192008879024,
-      "grad_norm": 1.0276777351490343,
+      "epoch": 345.0,
+      "grad_norm": 0.01684675867751184,
       "learning_rate": 1.582318458992267e-05,
-      "loss": 0.8701,
+      "loss": 0.0027,
       "step": 690
     },
     {
-      "epoch": 0.2571217166111728,
-      "grad_norm": 1.1308701537231916,
+      "epoch": 347.5,
+      "grad_norm": 0.032002543536054376,
       "learning_rate": 1.5506786103620146e-05,
-      "loss": 0.9146,
+      "loss": 0.0027,
       "step": 695
     },
     {
-      "epoch": 0.2589715131335553,
-      "grad_norm": 1.1220390082042924,
+      "epoch": 350.0,
+      "grad_norm": 0.021043035856906086,
       "learning_rate": 1.5193666442245402e-05,
-      "loss": 0.894,
+      "loss": 0.0027,
       "step": 700
     },
     {
-      "epoch": 0.26082130965593786,
-      "grad_norm": 0.9872150958917795,
+      "epoch": 352.5,
+      "grad_norm": 0.01601931955071107,
       "learning_rate": 1.4883911209587368e-05,
-      "loss": 0.8846,
+      "loss": 0.0026,
       "step": 705
     },
     {
-      "epoch": 0.2626711061783204,
-      "grad_norm": 1.0255612887775873,
+      "epoch": 355.0,
+      "grad_norm": 0.02601502051924795,
       "learning_rate": 1.4577605089633773e-05,
-      "loss": 0.8751,
+      "loss": 0.0026,
       "step": 710
     },
     {
-      "epoch": 0.26452090270070294,
-      "grad_norm": 1.0397487279123176,
+      "epoch": 357.5,
+      "grad_norm": 0.02710347701415514,
       "learning_rate": 1.427483182341936e-05,
-      "loss": 0.8553,
+      "loss": 0.0026,
       "step": 715
     },
     {
-      "epoch": 0.2663706992230855,
-      "grad_norm": 0.9811002147501025,
+      "epoch": 360.0,
+      "grad_norm": 0.03320644743977735,
       "learning_rate": 1.3975674186131937e-05,
-      "loss": 0.8811,
+      "loss": 0.0027,
       "step": 720
     },
     {
-      "epoch": 0.268220495745468,
-      "grad_norm": 0.9912452879027507,
+      "epoch": 362.5,
+      "grad_norm": 0.01591474795789328,
       "learning_rate": 1.3680213964482475e-05,
-      "loss": 0.882,
+      "loss": 0.0027,
       "step": 725
     },
     {
-      "epoch": 0.2700702922678505,
-      "grad_norm": 1.085598582193152,
+      "epoch": 365.0,
+      "grad_norm": 0.026155373607272086,
       "learning_rate": 1.3388531934345416e-05,
-      "loss": 0.8663,
+      "loss": 0.0027,
       "step": 730
     },
     {
-      "epoch": 0.27192008879023305,
-      "grad_norm": 1.0976908092393123,
+      "epoch": 367.5,
+      "grad_norm": 0.024271143128982743,
       "learning_rate": 1.3100707838675377e-05,
-      "loss": 0.8568,
+      "loss": 0.0026,
       "step": 735
     },
     {
-      "epoch": 0.2737698853126156,
-      "grad_norm": 1.110399579286176,
+      "epoch": 370.0,
+      "grad_norm": 0.016369678310535946,
       "learning_rate": 1.2816820365706206e-05,
-      "loss": 0.861,
+      "loss": 0.0027,
       "step": 740
     },
     {
-      "epoch": 0.27561968183499813,
-      "grad_norm": 0.9972527883665196,
+      "epoch": 372.5,
+      "grad_norm": 0.016955558044549856,
       "learning_rate": 1.2536947127438415e-05,
-      "loss": 0.8583,
+      "loss": 0.0026,
       "step": 745
     },
     {
-      "epoch": 0.27746947835738067,
-      "grad_norm": 0.9846092199040459,
+      "epoch": 375.0,
+      "grad_norm": 0.02239783849742695,
       "learning_rate": 1.2261164638420832e-05,
-      "loss": 0.8869,
+      "loss": 0.0026,
       "step": 750
     },
     {
-      "epoch": 0.2793192748797632,
-      "grad_norm": 1.0377584915303817,
+      "epoch": 377.5,
+      "grad_norm": 0.029924950679270815,
       "learning_rate": 1.198954829483227e-05,
-      "loss": 0.8689,
+      "loss": 0.0027,
       "step": 755
     },
     {
-      "epoch": 0.28116907140214575,
-      "grad_norm": 1.0080689260720659,
+      "epoch": 380.0,
+      "grad_norm": 0.028621427234193805,
       "learning_rate": 1.1722172353869008e-05,
-      "loss": 0.8494,
+      "loss": 0.0025,
       "step": 760
     },
     {
-      "epoch": 0.2830188679245283,
-      "grad_norm": 1.1377623343596663,
+      "epoch": 382.5,
+      "grad_norm": 0.02561136953891063,
       "learning_rate": 1.1459109913443567e-05,
-      "loss": 0.8789,
+      "loss": 0.0026,
       "step": 765
     },
     {
-      "epoch": 0.28486866444691084,
-      "grad_norm": 1.0844127152295038,
+      "epoch": 385.0,
+      "grad_norm": 0.028256041486332584,
       "learning_rate": 1.12004328922005e-05,
-      "loss": 0.8714,
+      "loss": 0.0027,
       "step": 770
     },
     {
-      "epoch": 0.2867184609692934,
-      "grad_norm": 0.9818291742586884,
+      "epoch": 387.5,
+      "grad_norm": 0.029729889846126924,
       "learning_rate": 1.094621200985454e-05,
-      "loss": 0.8762,
+      "loss": 0.0027,
       "step": 775
     },
     {
-      "epoch": 0.2885682574916759,
-      "grad_norm": 1.0561402094886105,
+      "epoch": 390.0,
+      "grad_norm": 0.014383243995550028,
       "learning_rate": 1.0696516767856546e-05,
-      "loss": 0.8602,
+      "loss": 0.0026,
       "step": 780
     },
     {
-      "epoch": 0.29041805401405846,
-      "grad_norm": 0.9507833621657122,
+      "epoch": 392.5,
+      "grad_norm": 0.015879736540587025,
       "learning_rate": 1.0451415430392474e-05,
-      "loss": 0.8493,
+      "loss": 0.0026,
       "step": 785
     },
     {
-      "epoch": 0.292267850536441,
-      "grad_norm": 1.0142419832874243,
+      "epoch": 395.0,
+      "grad_norm": 0.024629301532455877,
       "learning_rate": 1.0210975005720677e-05,
-      "loss": 0.8621,
+      "loss": 0.0026,
       "step": 790
     },
     {
-      "epoch": 0.29411764705882354,
-      "grad_norm": 1.0414075410194514,
+      "epoch": 397.5,
+      "grad_norm": 0.029707059150893382,
       "learning_rate": 9.975261227852472e-06,
-      "loss": 0.8867,
+      "loss": 0.0027,
       "step": 795
     },
     {
-      "epoch": 0.2959674435812061,
-      "grad_norm": 1.0172601224221116,
+      "epoch": 400.0,
+      "grad_norm": 0.0318537391592188,
       "learning_rate": 9.744338538581147e-06,
-      "loss": 0.861,
+      "loss": 0.0027,
       "step": 800
     },
     {
-      "epoch": 0.2978172401035886,
-      "grad_norm": 1.0520376994102576,
+      "epoch": 402.5,
+      "grad_norm": 0.017706015342844352,
       "learning_rate": 9.518270069864195e-06,
-      "loss": 0.8693,
+      "loss": 0.0027,
       "step": 805
     },
     {
-      "epoch": 0.29966703662597116,
-      "grad_norm": 1.0297669417176623,
+      "epoch": 405.0,
+      "grad_norm": 0.01724610650030409,
       "learning_rate": 9.297117626563687e-06,
-      "loss": 0.8556,
+      "loss": 0.0026,
       "step": 810
     },
     {
-      "epoch": 0.3015168331483537,
-      "grad_norm": 1.1288942237451884,
+      "epoch": 407.5,
+      "grad_norm": 0.018957851028085164,
       "learning_rate": 9.080941669549423e-06,
-      "loss": 0.8636,
+      "loss": 0.0026,
       "step": 815
     },
     {
-      "epoch": 0.30336662967073624,
-      "grad_norm": 1.0700035713287315,
+      "epoch": 410.0,
+      "grad_norm": 0.025418915769378568,
       "learning_rate": 8.86980129916955e-06,
-      "loss": 0.8735,
+      "loss": 0.0026,
       "step": 820
     },
     {
-      "epoch": 0.3052164261931188,
-      "grad_norm": 1.0433713434057506,
+      "epoch": 412.5,
+      "grad_norm": 0.01583223922193541,
       "learning_rate": 8.663754239093109e-06,
-      "loss": 0.8745,
+      "loss": 0.0027,
       "step": 825
     },
     {
-      "epoch": 0.30706622271550127,
-      "grad_norm": 0.9399395136256045,
+      "epoch": 415.0,
+      "grad_norm": 0.027164074218547888,
       "learning_rate": 8.462856820528952e-06,
-      "loss": 0.8625,
+      "loss": 0.0026,
       "step": 830
     },
     {
-      "epoch": 0.3089160192378838,
-      "grad_norm": 1.012084107494069,
+      "epoch": 417.5,
+      "grad_norm": 0.028647280916262126,
       "learning_rate": 8.26716396682537e-06,
-      "loss": 0.8575,
+      "loss": 0.0027,
       "step": 835
     },
     {
-      "epoch": 0.31076581576026635,
-      "grad_norm": 1.0479666608180018,
+      "epoch": 420.0,
+      "grad_norm": 0.019553123738546497,
       "learning_rate": 8.076729178454588e-06,
-      "loss": 0.8766,
+      "loss": 0.0026,
       "step": 840
     },
     {
-      "epoch": 0.3126156122826489,
-      "grad_norm": 1.137450242844205,
+      "epoch": 422.5,
+      "grad_norm": 0.025379025955769637,
       "learning_rate": 7.89160451838626e-06,
-      "loss": 0.8544,
+      "loss": 0.0027,
       "step": 845
     },
     {
-      "epoch": 0.31446540880503143,
-      "grad_norm": 1.0197993824595155,
+      "epoch": 425.0,
+      "grad_norm": 0.01750943138582632,
       "learning_rate": 7.711840597853998e-06,
-      "loss": 0.8583,
+      "loss": 0.0026,
       "step": 850
     },
     {
-      "epoch": 0.316315205327414,
-      "grad_norm": 1.0686381595688783,
+      "epoch": 427.5,
+      "grad_norm": 0.02520822601658372,
       "learning_rate": 7.537486562518735e-06,
-      "loss": 0.8518,
+      "loss": 0.0026,
       "step": 855
     },
     {
-      "epoch": 0.3181650018497965,
-      "grad_norm": 0.9978231071140236,
+      "epoch": 430.0,
+      "grad_norm": 0.01993484997782361,
       "learning_rate": 7.368590079032822e-06,
-      "loss": 0.8648,
+      "loss": 0.0026,
       "step": 860
     },
     {
-      "epoch": 0.32001479837217905,
-      "grad_norm": 1.0560393356842295,
+      "epoch": 432.5,
+      "grad_norm": 0.019158523070010834,
       "learning_rate": 7.205197322008425e-06,
-      "loss": 0.8349,
+      "loss": 0.0027,
       "step": 865
     },
     {
-      "epoch": 0.3218645948945616,
-      "grad_norm": 0.9842318437238298,
+      "epoch": 435.0,
+      "grad_norm": 0.015239850907125138,
       "learning_rate": 7.047352961393844e-06,
-      "loss": 0.8595,
+      "loss": 0.0027,
       "step": 870
     },
     {
-      "epoch": 0.32371439141694414,
-      "grad_norm": 0.9952118512469338,
+      "epoch": 437.5,
+      "grad_norm": 0.020559758942860218,
       "learning_rate": 6.8951001502612065e-06,
-      "loss": 0.869,
+      "loss": 0.0027,
       "step": 875
     },
     {
-      "epoch": 0.3255641879393267,
-      "grad_norm": 0.9424925973460067,
+      "epoch": 440.0,
+      "grad_norm": 0.02381279446911383,
       "learning_rate": 6.748480513008844e-06,
-      "loss": 0.8745,
+      "loss": 0.0025,
       "step": 880
     },
     {
-      "epoch": 0.3274139844617092,
-      "grad_norm": 1.0535831291700837,
+      "epoch": 442.5,
+      "grad_norm": 0.014589626300082677,
       "learning_rate": 6.607534133981594e-06,
-      "loss": 0.8748,
+      "loss": 0.0026,
       "step": 885
     },
     {
-      "epoch": 0.32926378098409176,
-      "grad_norm": 1.0723567866928634,
+      "epoch": 445.0,
+      "grad_norm": 0.026727121558605703,
       "learning_rate": 6.472299546512134e-06,
-      "loss": 0.8815,
+      "loss": 0.0026,
       "step": 890
     },
     {
-      "epoch": 0.3311135775064743,
-      "grad_norm": 1.0482717572477003,
+      "epoch": 447.5,
+      "grad_norm": 0.01580348324753083,
       "learning_rate": 6.342813722386374e-06,
-      "loss": 0.8543,
+      "loss": 0.0027,
       "step": 895
     },
     {
-      "epoch": 0.33296337402885684,
-      "grad_norm": 1.046401991122096,
+      "epoch": 450.0,
+      "grad_norm": 0.02334311857794958,
       "learning_rate": 6.219112061735721e-06,
-      "loss": 0.8785,
+      "loss": 0.0026,
       "step": 900
     },
     {
-      "epoch": 0.3348131705512394,
-      "grad_norm": 1.0211208893623536,
+      "epoch": 452.5,
+      "grad_norm": 0.022995449553396018,
       "learning_rate": 6.1012283833590465e-06,
-      "loss": 0.8757,
+      "loss": 0.0026,
       "step": 905
     },
     {
-      "epoch": 0.3366629670736219,
-      "grad_norm": 1.0035173287590444,
+      "epoch": 455.0,
+      "grad_norm": 0.03391053587500504,
       "learning_rate": 5.989194915476954e-06,
-      "loss": 0.8442,
+      "loss": 0.0027,
       "step": 910
     },
     {
-      "epoch": 0.33851276359600446,
-      "grad_norm": 0.9878066592821019,
+      "epoch": 457.5,
+      "grad_norm": 0.0281194121062883,
       "learning_rate": 5.883042286920918e-06,
-      "loss": 0.8638,
+      "loss": 0.0025,
       "step": 915
     },
     {
-      "epoch": 0.340362560118387,
-      "grad_norm": 1.05735602894325,
+      "epoch": 460.0,
+      "grad_norm": 0.018983581090637,
       "learning_rate": 5.782799518759658e-06,
-      "loss": 0.8593,
+      "loss": 0.0025,
       "step": 920
     },
     {
-      "epoch": 0.3422123566407695,
-      "grad_norm": 1.0649111496758787,
+      "epoch": 462.5,
+      "grad_norm": 0.024476355758836905,
       "learning_rate": 5.688494016365067e-06,
-      "loss": 0.845,
+      "loss": 0.0026,
       "step": 925
     },
     {
-      "epoch": 0.34406215316315203,
-      "grad_norm": 0.957225274905496,
+      "epoch": 465.0,
+      "grad_norm": 0.023061048781595993,
       "learning_rate": 5.600151561919871e-06,
-      "loss": 0.8134,
+      "loss": 0.0027,
       "step": 930
     },
     {
-      "epoch": 0.34591194968553457,
-      "grad_norm": 1.0690464205878594,
+      "epoch": 467.5,
+      "grad_norm": 0.012496893287463113,
       "learning_rate": 5.517796307369017e-06,
-      "loss": 0.8364,
+      "loss": 0.0026,
       "step": 935
     },
     {
-      "epoch": 0.3477617462079171,
-      "grad_norm": 1.0520479367896434,
+      "epoch": 470.0,
+      "grad_norm": 0.01991683594417727,
       "learning_rate": 5.44145076781679e-06,
-      "loss": 0.8697,
+      "loss": 0.0026,
       "step": 940
     },
     {
-      "epoch": 0.34961154273029965,
-      "grad_norm": 0.9648256295717741,
+      "epoch": 472.5,
+      "grad_norm": 0.028037001206555136,
       "learning_rate": 5.371135815371398e-06,
-      "loss": 0.8334,
+      "loss": 0.0027,
       "step": 945
     },
     {
-      "epoch": 0.3514613392526822,
-      "grad_norm": 1.0090319357515105,
+      "epoch": 475.0,
+      "grad_norm": 0.026283232329898983,
       "learning_rate": 5.3068706734387484e-06,
-      "loss": 0.866,
+      "loss": 0.0026,
       "step": 950
     },
     {
-      "epoch": 0.35331113577506473,
-      "grad_norm": 1.0792234217081147,
+      "epoch": 477.5,
+      "grad_norm": 0.026815406305102027,
       "learning_rate": 5.248672911466959e-06,
-      "loss": 0.8796,
+      "loss": 0.0026,
       "step": 955
     },
     {
-      "epoch": 0.3551609322974473,
-      "grad_norm": 0.9703249482629162,
+      "epoch": 480.0,
+      "grad_norm": 0.01847886314594556,
       "learning_rate": 5.196558440143059e-06,
-      "loss": 0.8329,
+      "loss": 0.0025,
       "step": 960
     },
     {
-      "epoch": 0.3570107288198298,
-      "grad_norm": 1.0041999264211245,
+      "epoch": 482.5,
+      "grad_norm": 0.019370493214584474,
       "learning_rate": 5.150541507043143e-06,
-      "loss": 0.8871,
+      "loss": 0.0027,
       "step": 965
     },
     {
-      "epoch": 0.35886052534221236,
-      "grad_norm": 0.9816980969927207,
+      "epoch": 485.0,
+      "grad_norm": 0.021779585772489417,
       "learning_rate": 5.110634692737244e-06,
-      "loss": 0.8391,
+      "loss": 0.0026,
       "step": 970
     },
     {
-      "epoch": 0.3607103218645949,
-      "grad_norm": 0.9903601458320098,
+      "epoch": 487.5,
+      "grad_norm": 0.02541037015533656,
       "learning_rate": 5.0768489073499295e-06,
-      "loss": 0.8707,
+      "loss": 0.0027,
       "step": 975
     },
     {
-      "epoch": 0.36256011838697744,
-      "grad_norm": 0.9605358480816072,
+      "epoch": 490.0,
+      "grad_norm": 0.02945361459676501,
       "learning_rate": 5.049193387577574e-06,
-      "loss": 0.8677,
+      "loss": 0.0026,
       "step": 980
     },
     {
-      "epoch": 0.36440991490936,
-      "grad_norm": 1.0149824944311472,
+      "epoch": 492.5,
+      "grad_norm": 0.017922772556181753,
       "learning_rate": 5.027675694163144e-06,
-      "loss": 0.8252,
+      "loss": 0.0028,
       "step": 985
     },
     {
-      "epoch": 0.3662597114317425,
-      "grad_norm": 1.0514282882320594,
+      "epoch": 495.0,
+      "grad_norm": 0.023023048365518207,
       "learning_rate": 5.012301709829164e-06,
-      "loss": 0.8361,
+      "loss": 0.0026,
       "step": 990
     },
     {
-      "epoch": 0.36810950795412506,
-      "grad_norm": 1.0144426282975678,
+      "epoch": 497.5,
+      "grad_norm": 0.020646776127788924,
       "learning_rate": 5.003075637669448e-06,
-      "loss": 0.8604,
+      "loss": 0.0027,
       "step": 995
     },
     {
-      "epoch": 0.3699593044765076,
-      "grad_norm": 1.041222716272992,
+      "epoch": 500.0,
+      "grad_norm": 0.0148034159233786,
       "learning_rate": 5e-06,
-      "loss": 0.8502,
+      "loss": 0.0027,
       "step": 1000
     },
     {
-      "epoch": 0.3699593044765076,
+      "epoch": 500.0,
       "step": 1000,
-      "total_flos": 28487712768000.0,
-      "train_loss": 0.920774197101593,
-      "train_runtime": 1759.2244,
-      "train_samples_per_second": 18.19,
-      "train_steps_per_second": 0.568
+      "total_flos": 28457722773504.0,
+      "train_loss": 0.008928275344893336,
+      "train_runtime": 1701.2794,
+      "train_samples_per_second": 18.809,
+      "train_steps_per_second": 0.588
     }
   ],
   "logging_steps": 5,
   "max_steps": 1000,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 1,
+  "num_train_epochs": 500,
   "save_steps": 100,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -1435,7 +1435,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 28487712768000.0,
+  "total_flos": 28457722773504.0,
   "train_batch_size": 16,
   "trial_name": null,
   "trial_params": null