End of training

README.md

@@ -4,6 +4,7 @@ license: llama3.1
 base_model: meta-llama/Llama-3.1-8B
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: OH_original_wo_airoboros
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # OH_original_wo_airoboros
 
-This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the mlfoundations-dev/OH_original_wo_airoboros dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6065
 

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 2.993050193050193,
-    "eval_loss": 0.611685574054718,
-    "eval_runtime": 175.2261,
-    "eval_samples_per_second": 49.764,
-    "eval_steps_per_second": 0.394,
+    "eval_loss": 0.6065478920936584,
+    "eval_runtime": 174.4122,
+    "eval_samples_per_second": 49.997,
+    "eval_steps_per_second": 0.396,
     "total_flos": 1622692331520000.0,
-    "train_loss": 0.5660572312810719,
-    "train_runtime": 29124.935,
-    "train_samples_per_second": 17.065,
+    "train_loss": 0.5687421354839061,
+    "train_runtime": 29217.0698,
+    "train_samples_per_second": 17.012,
     "train_steps_per_second": 0.033
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 2.993050193050193,
-    "eval_loss": 0.611685574054718,
-    "eval_runtime": 175.2261,
-    "eval_samples_per_second": 49.764,
-    "eval_steps_per_second": 0.394
+    "eval_loss": 0.6065478920936584,
+    "eval_runtime": 174.4122,
+    "eval_samples_per_second": 49.997,
+    "eval_steps_per_second": 0.396
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.993050193050193,
     "total_flos": 1622692331520000.0,
-    "train_loss": 0.5660572312810719,
-    "train_runtime": 29124.935,
-    "train_samples_per_second": 17.065,
+    "train_loss": 0.5687421354839061,
+    "train_runtime": 29217.0698,
+    "train_samples_per_second": 17.012,
     "train_steps_per_second": 0.033
 }

trainer_state.json

@@ -10,697 +10,697 @@
   "log_history": [
     {
       "epoch": 0.03088803088803089,
-      "grad_norm": 109.2129922948738,
+      "grad_norm": 1.9920737405210878,
       "learning_rate": 5e-06,
-      "loss": 0.8218,
+      "loss": 0.8015,
       "step": 10
     },
     {
       "epoch": 0.06177606177606178,
-      "grad_norm": 1.5509789288186628,
+      "grad_norm": 2.6204521466523314,
       "learning_rate": 5e-06,
-      "loss": 0.7505,
+      "loss": 0.7199,
       "step": 20
     },
     {
       "epoch": 0.09266409266409266,
-      "grad_norm": 1.0237535334293637,
+      "grad_norm": 1.3723889792780992,
       "learning_rate": 5e-06,
-      "loss": 0.7022,
+      "loss": 0.6903,
       "step": 30
     },
     {
       "epoch": 0.12355212355212356,
-      "grad_norm": 0.9555198347715455,
+      "grad_norm": 0.8864238330359895,
       "learning_rate": 5e-06,
-      "loss": 0.6862,
+      "loss": 0.6793,
       "step": 40
     },
     {
       "epoch": 0.15444015444015444,
-      "grad_norm": 0.9962768534382307,
+      "grad_norm": 13.397151069895909,
       "learning_rate": 5e-06,
-      "loss": 0.6721,
+      "loss": 0.6675,
       "step": 50
     },
     {
       "epoch": 0.18532818532818532,
-      "grad_norm": 1.7530907194148484,
+      "grad_norm": 2.1374873857216876,
       "learning_rate": 5e-06,
-      "loss": 0.6509,
+      "loss": 0.6488,
       "step": 60
     },
     {
       "epoch": 0.21621621621621623,
-      "grad_norm": 0.7959383217983834,
+      "grad_norm": 0.7775938162830551,
       "learning_rate": 5e-06,
-      "loss": 0.6472,
+      "loss": 0.6457,
       "step": 70
     },
     {
       "epoch": 0.2471042471042471,
-      "grad_norm": 0.7944327039168986,
+      "grad_norm": 0.8956285245200575,
       "learning_rate": 5e-06,
-      "loss": 0.65,
+      "loss": 0.6479,
       "step": 80
     },
     {
       "epoch": 0.277992277992278,
-      "grad_norm": 0.6922868471685667,
+      "grad_norm": 0.7136145785303137,
       "learning_rate": 5e-06,
-      "loss": 0.639,
+      "loss": 0.6365,
       "step": 90
     },
     {
       "epoch": 0.3088803088803089,
-      "grad_norm": 0.6117044147831933,
+      "grad_norm": 0.5386743245333256,
       "learning_rate": 5e-06,
-      "loss": 0.6398,
+      "loss": 0.6366,
       "step": 100
     },
     {
       "epoch": 0.33976833976833976,
-      "grad_norm": 0.6349747322397297,
+      "grad_norm": 0.5331290951707172,
       "learning_rate": 5e-06,
-      "loss": 0.6335,
+      "loss": 0.6305,
       "step": 110
     },
     {
       "epoch": 0.37065637065637064,
-      "grad_norm": 0.6236670890266053,
+      "grad_norm": 0.5589168290698243,
       "learning_rate": 5e-06,
-      "loss": 0.6337,
+      "loss": 0.6305,
       "step": 120
     },
     {
       "epoch": 0.4015444015444015,
-      "grad_norm": 0.6724570938621872,
+      "grad_norm": 0.6427317993478433,
       "learning_rate": 5e-06,
-      "loss": 0.6255,
+      "loss": 0.6221,
       "step": 130
     },
     {
       "epoch": 0.43243243243243246,
-      "grad_norm": 0.8157933757633341,
+      "grad_norm": 0.5345254165110678,
       "learning_rate": 5e-06,
-      "loss": 0.6283,
+      "loss": 0.6251,
       "step": 140
     },
     {
       "epoch": 0.46332046332046334,
-      "grad_norm": 0.5839495663310666,
+      "grad_norm": 0.5523122802530471,
       "learning_rate": 5e-06,
-      "loss": 0.6214,
+      "loss": 0.6183,
       "step": 150
     },
     {
       "epoch": 0.4942084942084942,
-      "grad_norm": 0.5048909710838132,
+      "grad_norm": 0.6345828990779662,
       "learning_rate": 5e-06,
-      "loss": 0.6243,
+      "loss": 0.6211,
       "step": 160
     },
     {
       "epoch": 0.525096525096525,
-      "grad_norm": 0.8095059876811945,
+      "grad_norm": 0.5635690118220865,
       "learning_rate": 5e-06,
-      "loss": 0.6251,
+      "loss": 0.6222,
       "step": 170
     },
     {
       "epoch": 0.555984555984556,
-      "grad_norm": 0.5136815801780633,
+      "grad_norm": 0.5674431409857721,
       "learning_rate": 5e-06,
-      "loss": 0.6145,
+      "loss": 0.6121,
       "step": 180
     },
     {
       "epoch": 0.5868725868725869,
-      "grad_norm": 0.6364035663779027,
+      "grad_norm": 0.9846408481503562,
       "learning_rate": 5e-06,
-      "loss": 0.6298,
+      "loss": 0.627,
       "step": 190
     },
     {
       "epoch": 0.6177606177606177,
-      "grad_norm": 0.6844817393850995,
+      "grad_norm": 0.5946873633912808,
       "learning_rate": 5e-06,
-      "loss": 0.6197,
+      "loss": 0.6171,
       "step": 200
     },
     {
       "epoch": 0.6486486486486487,
-      "grad_norm": 0.7483950060039181,
+      "grad_norm": 1.1118602099699486,
       "learning_rate": 5e-06,
-      "loss": 0.6269,
+      "loss": 0.6244,
       "step": 210
     },
     {
       "epoch": 0.6795366795366795,
-      "grad_norm": 0.7350511698635498,
+      "grad_norm": 0.6560189254340667,
       "learning_rate": 5e-06,
-      "loss": 0.6175,
+      "loss": 0.615,
       "step": 220
     },
     {
       "epoch": 0.7104247104247104,
-      "grad_norm": 0.6246522556489467,
+      "grad_norm": 1.108952229503608,
       "learning_rate": 5e-06,
-      "loss": 0.6172,
+      "loss": 0.6147,
       "step": 230
     },
     {
       "epoch": 0.7413127413127413,
-      "grad_norm": 0.5845884993818068,
+      "grad_norm": 0.6239111214970273,
       "learning_rate": 5e-06,
-      "loss": 0.6119,
+      "loss": 0.6097,
       "step": 240
     },
     {
       "epoch": 0.7722007722007722,
-      "grad_norm": 0.6246906999878916,
+      "grad_norm": 1.0310044782453138,
       "learning_rate": 5e-06,
-      "loss": 0.6191,
+      "loss": 0.6167,
       "step": 250
     },
     {
       "epoch": 0.803088803088803,
-      "grad_norm": 0.5725768617907299,
+      "grad_norm": 0.5083008941812818,
       "learning_rate": 5e-06,
-      "loss": 0.6111,
+      "loss": 0.6088,
       "step": 260
     },
     {
       "epoch": 0.833976833976834,
-      "grad_norm": 0.7575125169415168,
+      "grad_norm": 0.5393819554919369,
       "learning_rate": 5e-06,
-      "loss": 0.6078,
+      "loss": 0.6054,
       "step": 270
     },
     {
       "epoch": 0.8648648648648649,
-      "grad_norm": 0.6292537869493213,
+      "grad_norm": 0.501525790082182,
       "learning_rate": 5e-06,
-      "loss": 0.6119,
+      "loss": 0.6095,
       "step": 280
     },
     {
       "epoch": 0.8957528957528957,
-      "grad_norm": 0.6613666650568315,
+      "grad_norm": 0.5189867584989447,
       "learning_rate": 5e-06,
-      "loss": 0.6219,
+      "loss": 0.6199,
       "step": 290
     },
     {
       "epoch": 0.9266409266409267,
-      "grad_norm": 0.482392770012809,
+      "grad_norm": 0.4557994704814137,
       "learning_rate": 5e-06,
-      "loss": 0.6032,
+      "loss": 0.6008,
       "step": 300
     },
     {
       "epoch": 0.9575289575289575,
-      "grad_norm": 0.5691595473530817,
+      "grad_norm": 0.5100750571195128,
       "learning_rate": 5e-06,
-      "loss": 0.6137,
+      "loss": 0.6118,
       "step": 310
     },
     {
       "epoch": 0.9884169884169884,
-      "grad_norm": 0.5605696956503327,
+      "grad_norm": 0.5597550458388434,
       "learning_rate": 5e-06,
-      "loss": 0.6059,
+      "loss": 0.604,
       "step": 320
     },
     {
       "epoch": 0.9976833976833976,
-      "eval_loss": 0.611504316329956,
-      "eval_runtime": 174.6046,
-      "eval_samples_per_second": 49.941,
-      "eval_steps_per_second": 0.395,
+      "eval_loss": 0.6096363663673401,
+      "eval_runtime": 174.3952,
+      "eval_samples_per_second": 50.001,
+      "eval_steps_per_second": 0.396,
       "step": 323
     },
     {
       "epoch": 1.0193050193050193,
-      "grad_norm": 0.7653552774499555,
+      "grad_norm": 1.0059049758489014,
       "learning_rate": 5e-06,
-      "loss": 0.577,
+      "loss": 0.5788,
       "step": 330
     },
     {
       "epoch": 1.05019305019305,
-      "grad_norm": 0.6717959962039358,
+      "grad_norm": 0.8266294447843203,
       "learning_rate": 5e-06,
-      "loss": 0.5516,
+      "loss": 0.5555,
       "step": 340
     },
     {
       "epoch": 1.0810810810810811,
-      "grad_norm": 0.7810736772266839,
+      "grad_norm": 0.6817740146761346,
       "learning_rate": 5e-06,
-      "loss": 0.5527,
+      "loss": 0.5568,
       "step": 350
     },
     {
       "epoch": 1.111969111969112,
-      "grad_norm": 0.5222987434687074,
+      "grad_norm": 0.5212640526638012,
       "learning_rate": 5e-06,
-      "loss": 0.5672,
+      "loss": 0.5711,
       "step": 360
     },
     {
       "epoch": 1.1428571428571428,
-      "grad_norm": 0.6086528109996991,
+      "grad_norm": 0.6082293357332792,
       "learning_rate": 5e-06,
-      "loss": 0.559,
+      "loss": 0.5631,
       "step": 370
     },
     {
       "epoch": 1.1737451737451738,
-      "grad_norm": 0.5603581613184462,
+      "grad_norm": 0.4984378073822691,
       "learning_rate": 5e-06,
-      "loss": 0.5557,
+      "loss": 0.5597,
       "step": 380
     },
     {
       "epoch": 1.2046332046332047,
-      "grad_norm": 0.5733679682136087,
+      "grad_norm": 0.5400232508156533,
       "learning_rate": 5e-06,
-      "loss": 0.5523,
+      "loss": 0.5563,
       "step": 390
     },
     {
       "epoch": 1.2355212355212355,
-      "grad_norm": 0.5892263714928079,
+      "grad_norm": 0.4863797369607136,
       "learning_rate": 5e-06,
-      "loss": 0.5581,
+      "loss": 0.563,
       "step": 400
     },
     {
       "epoch": 1.2664092664092665,
-      "grad_norm": 0.5383912423961117,
+      "grad_norm": 0.5275923409775821,
       "learning_rate": 5e-06,
-      "loss": 0.5587,
+      "loss": 0.5625,
       "step": 410
     },
     {
       "epoch": 1.2972972972972974,
-      "grad_norm": 0.5266690487444351,
+      "grad_norm": 0.527704404092815,
       "learning_rate": 5e-06,
-      "loss": 0.5538,
+      "loss": 0.5577,
       "step": 420
     },
     {
       "epoch": 1.3281853281853282,
-      "grad_norm": 0.6329004292969694,
+      "grad_norm": 0.6166415284725348,
       "learning_rate": 5e-06,
-      "loss": 0.5621,
+      "loss": 0.5658,
       "step": 430
     },
     {
       "epoch": 1.359073359073359,
-      "grad_norm": 0.7361457930766753,
+      "grad_norm": 0.6544026128628749,
       "learning_rate": 5e-06,
-      "loss": 0.5507,
+      "loss": 0.5544,
       "step": 440
     },
     {
       "epoch": 1.3899613899613898,
-      "grad_norm": 0.5514849953088692,
+      "grad_norm": 0.5151291141077943,
       "learning_rate": 5e-06,
-      "loss": 0.5458,
+      "loss": 0.5497,
       "step": 450
     },
     {
       "epoch": 1.420849420849421,
-      "grad_norm": 0.6157873717956057,
+      "grad_norm": 0.5534465037275664,
       "learning_rate": 5e-06,
-      "loss": 0.5576,
+      "loss": 0.5614,
       "step": 460
     },
     {
       "epoch": 1.4517374517374517,
-      "grad_norm": 0.6133558219166387,
+      "grad_norm": 0.5300403375821853,
       "learning_rate": 5e-06,
-      "loss": 0.5549,
+      "loss": 0.5583,
       "step": 470
     },
     {
       "epoch": 1.4826254826254825,
-      "grad_norm": 0.8739126560777261,
+      "grad_norm": 0.7438189920291365,
       "learning_rate": 5e-06,
-      "loss": 0.5544,
+      "loss": 0.558,
       "step": 480
     },
     {
       "epoch": 1.5135135135135136,
-      "grad_norm": 0.6361501246308614,
+      "grad_norm": 0.531280818371624,
       "learning_rate": 5e-06,
-      "loss": 0.5593,
+      "loss": 0.5629,
       "step": 490
     },
     {
       "epoch": 1.5444015444015444,
-      "grad_norm": 0.5962865704485671,
+      "grad_norm": 0.5680517069700968,
       "learning_rate": 5e-06,
-      "loss": 0.5536,
+      "loss": 0.557,
       "step": 500
     },
     {
       "epoch": 1.5752895752895753,
-      "grad_norm": 0.5717477466618819,
+      "grad_norm": 0.5612219075386209,
       "learning_rate": 5e-06,
-      "loss": 0.5613,
+      "loss": 0.5653,
       "step": 510
     },
     {
       "epoch": 1.606177606177606,
-      "grad_norm": 0.66303577521287,
+      "grad_norm": 0.649010796286653,
       "learning_rate": 5e-06,
-      "loss": 0.5515,
+      "loss": 0.5553,
       "step": 520
     },
     {
       "epoch": 1.637065637065637,
-      "grad_norm": 0.567568871578932,
+      "grad_norm": 0.5724021305893356,
       "learning_rate": 5e-06,
-      "loss": 0.552,
+      "loss": 0.5554,
       "step": 530
     },
     {
       "epoch": 1.667953667953668,
-      "grad_norm": 0.6260947334671597,
+      "grad_norm": 0.5266418342750984,
       "learning_rate": 5e-06,
-      "loss": 0.5523,
+      "loss": 0.556,
       "step": 540
     },
     {
       "epoch": 1.698841698841699,
-      "grad_norm": 0.5443908156986677,
+      "grad_norm": 0.4764947467562162,
       "learning_rate": 5e-06,
-      "loss": 0.5597,
+      "loss": 0.5637,
       "step": 550
     },
     {
       "epoch": 1.7297297297297298,
-      "grad_norm": 0.5875127617844319,
+      "grad_norm": 0.49367950454054643,
       "learning_rate": 5e-06,
-      "loss": 0.5577,
+      "loss": 0.5616,
       "step": 560
     },
     {
       "epoch": 1.7606177606177607,
-      "grad_norm": 0.5839768971343271,
+      "grad_norm": 0.48282246019994013,
       "learning_rate": 5e-06,
-      "loss": 0.5556,
+      "loss": 0.559,
       "step": 570
     },
     {
       "epoch": 1.7915057915057915,
-      "grad_norm": 0.5415996097268374,
+      "grad_norm": 0.47502284833211744,
       "learning_rate": 5e-06,
-      "loss": 0.5541,
+      "loss": 0.5575,
       "step": 580
     },
     {
       "epoch": 1.8223938223938223,
-      "grad_norm": 0.5047758992113026,
+      "grad_norm": 0.45633738284447206,
       "learning_rate": 5e-06,
-      "loss": 0.5498,
+      "loss": 0.5532,
       "step": 590
     },
     {
       "epoch": 1.8532818532818531,
-      "grad_norm": 0.6095158836516334,
+      "grad_norm": 0.5759433055884126,
       "learning_rate": 5e-06,
-      "loss": 0.5577,
+      "loss": 0.5615,
       "step": 600
     },
     {
       "epoch": 1.8841698841698842,
-      "grad_norm": 0.6179499758866378,
+      "grad_norm": 0.5858108727658949,
       "learning_rate": 5e-06,
-      "loss": 0.5561,
+      "loss": 0.5593,
       "step": 610
     },
     {
       "epoch": 1.915057915057915,
-      "grad_norm": 0.6168326581011625,
+      "grad_norm": 0.5395422279987274,
       "learning_rate": 5e-06,
-      "loss": 0.5623,
+      "loss": 0.566,
       "step": 620
     },
     {
       "epoch": 1.945945945945946,
-      "grad_norm": 0.5503294354981652,
+      "grad_norm": 0.544007523438176,
       "learning_rate": 5e-06,
-      "loss": 0.5513,
+      "loss": 0.5546,
       "step": 630
     },
     {
       "epoch": 1.9768339768339769,
-      "grad_norm": 0.5514628691685661,
+      "grad_norm": 0.4756136721421921,
       "learning_rate": 5e-06,
-      "loss": 0.5543,
+      "loss": 0.5577,
       "step": 640
     },
     {
       "epoch": 1.9984555984555985,
-      "eval_loss": 0.6036180853843689,
-      "eval_runtime": 175.7287,
-      "eval_samples_per_second": 49.622,
+      "eval_loss": 0.6016931533813477,
+      "eval_runtime": 175.4937,
+      "eval_samples_per_second": 49.688,
       "eval_steps_per_second": 0.393,
       "step": 647
     },
     {
       "epoch": 2.0077220077220077,
-      "grad_norm": 1.1966714312564959,
+      "grad_norm": 0.9647134139465964,
       "learning_rate": 5e-06,
-      "loss": 0.5432,
+      "loss": 0.5481,
       "step": 650
     },
     {
       "epoch": 2.0386100386100385,
-      "grad_norm": 0.6689403789809841,
+      "grad_norm": 0.6126014654142371,
       "learning_rate": 5e-06,
-      "loss": 0.5094,
+      "loss": 0.5195,
       "step": 660
     },
     {
       "epoch": 2.0694980694980694,
-      "grad_norm": 0.61180285844113,
+      "grad_norm": 0.642576302083791,
       "learning_rate": 5e-06,
-      "loss": 0.5053,
+      "loss": 0.5154,
       "step": 670
     },
     {
       "epoch": 2.1003861003861,
-      "grad_norm": 0.6066089470745547,
+      "grad_norm": 0.6053831547491819,
       "learning_rate": 5e-06,
-      "loss": 0.5038,
+      "loss": 0.5141,
       "step": 680
     },
     {
       "epoch": 2.1312741312741315,
-      "grad_norm": 0.7339359470891843,
+      "grad_norm": 0.570602965823511,
       "learning_rate": 5e-06,
-      "loss": 0.4866,
+      "loss": 0.4963,
       "step": 690
     },
     {
       "epoch": 2.1621621621621623,
-      "grad_norm": 0.6280327917691826,
+      "grad_norm": 0.5789932913521146,
       "learning_rate": 5e-06,
-      "loss": 0.4925,
+      "loss": 0.5023,
       "step": 700
     },
     {
       "epoch": 2.193050193050193,
-      "grad_norm": 0.6228681526915294,
+      "grad_norm": 0.49902888045361504,
       "learning_rate": 5e-06,
-      "loss": 0.4964,
+      "loss": 0.5053,
       "step": 710
     },
     {
       "epoch": 2.223938223938224,
-      "grad_norm": 0.7409827598226171,
+      "grad_norm": 0.5200818247457489,
       "learning_rate": 5e-06,
-      "loss": 0.4991,
+      "loss": 0.5086,
       "step": 720
     },
     {
       "epoch": 2.2548262548262548,
-      "grad_norm": 0.5789864024622833,
+      "grad_norm": 0.5147752817966699,
       "learning_rate": 5e-06,
-      "loss": 0.5047,
+      "loss": 0.5144,
       "step": 730
     },
     {
       "epoch": 2.2857142857142856,
-      "grad_norm": 0.6184054511523953,
+      "grad_norm": 0.5202904589332674,
       "learning_rate": 5e-06,
-      "loss": 0.4983,
+      "loss": 0.5075,
       "step": 740
     },
     {
       "epoch": 2.3166023166023164,
-      "grad_norm": 0.7218033338604144,
+      "grad_norm": 0.6419271084211798,
       "learning_rate": 5e-06,
-      "loss": 0.5006,
+      "loss": 0.5095,
       "step": 750
     },
     {
       "epoch": 2.3474903474903477,
-      "grad_norm": 0.5488090893995611,
+      "grad_norm": 0.48888244575595774,
       "learning_rate": 5e-06,
-      "loss": 0.4982,
+      "loss": 0.5074,
       "step": 760
     },
     {
       "epoch": 2.3783783783783785,
-      "grad_norm": 0.7369531886408642,
+      "grad_norm": 0.4949992881722656,
       "learning_rate": 5e-06,
-      "loss": 0.4967,
+      "loss": 0.5058,
       "step": 770
     },
     {
       "epoch": 2.4092664092664093,
-      "grad_norm": 0.5519469047632208,
+      "grad_norm": 0.6977527327479829,
       "learning_rate": 5e-06,
-      "loss": 0.4965,
+      "loss": 0.5055,
       "step": 780
     },
     {
       "epoch": 2.44015444015444,
-      "grad_norm": 0.6584775748338472,
+      "grad_norm": 0.5281157241550238,
       "learning_rate": 5e-06,
-      "loss": 0.501,
+      "loss": 0.5101,
       "step": 790
     },
     {
       "epoch": 2.471042471042471,
-      "grad_norm": 0.6589409942831427,
+      "grad_norm": 0.5317168799948615,
       "learning_rate": 5e-06,
-      "loss": 0.5006,
+      "loss": 0.5092,
       "step": 800
     },
     {
       "epoch": 2.501930501930502,
-      "grad_norm": 0.6035656966090169,
+      "grad_norm": 0.5665091196485048,
       "learning_rate": 5e-06,
-      "loss": 0.504,
+      "loss": 0.5123,
       "step": 810
     },
     {
       "epoch": 2.532818532818533,
-      "grad_norm": 0.5939046793612662,
+      "grad_norm": 0.5733443367920803,
       "learning_rate": 5e-06,
-      "loss": 0.4962,
+      "loss": 0.5045,
       "step": 820
     },
     {
       "epoch": 2.563706563706564,
-      "grad_norm": 0.6332349300931522,
+      "grad_norm": 0.4895951656550531,
       "learning_rate": 5e-06,
-      "loss": 0.5184,
+      "loss": 0.5245,
       "step": 830
     },
     {
       "epoch": 2.5945945945945947,
-      "grad_norm": 0.5980197287456133,
+      "grad_norm": 0.5509555644837258,
       "learning_rate": 5e-06,
-      "loss": 0.4989,
+      "loss": 0.5074,
       "step": 840
     },
     {
       "epoch": 2.6254826254826256,
-      "grad_norm": 0.5239673559847541,
+      "grad_norm": 0.5242119378254794,
       "learning_rate": 5e-06,
-      "loss": 0.5013,
+      "loss": 0.5102,
       "step": 850
     },
     {
       "epoch": 2.6563706563706564,
-      "grad_norm": 0.6161550811061723,
+      "grad_norm": 0.6137964577932209,
       "learning_rate": 5e-06,
-      "loss": 0.4979,
+      "loss": 0.5066,
       "step": 860
     },
     {
       "epoch": 2.687258687258687,
-      "grad_norm": 0.5658973017035395,
+      "grad_norm": 0.517696061441138,
       "learning_rate": 5e-06,
-      "loss": 0.5069,
+      "loss": 0.5156,
       "step": 870
     },
     {
       "epoch": 2.718146718146718,
-      "grad_norm": 0.5709137558646814,
+      "grad_norm": 0.5449458713675641,
       "learning_rate": 5e-06,
-      "loss": 0.5065,
+      "loss": 0.5153,
       "step": 880
     },
     {
       "epoch": 2.749034749034749,
-      "grad_norm": 0.6389113703934911,
+      "grad_norm": 0.5127690582430875,
       "learning_rate": 5e-06,
-      "loss": 0.5037,
+      "loss": 0.5121,
       "step": 890
     },
     {
       "epoch": 2.7799227799227797,
-      "grad_norm": 0.6445028760804165,
+      "grad_norm": 0.5989390375727383,
       "learning_rate": 5e-06,
-      "loss": 0.5064,
+      "loss": 0.5149,
       "step": 900
     },
     {
       "epoch": 2.810810810810811,
-      "grad_norm": 0.6013274832800054,
+      "grad_norm": 0.5398603643342522,
       "learning_rate": 5e-06,
-      "loss": 0.5032,
+      "loss": 0.5116,
       "step": 910
     },
     {
       "epoch": 2.841698841698842,
-      "grad_norm": 0.5664869441381479,
+      "grad_norm": 0.5257472179121192,
       "learning_rate": 5e-06,
-      "loss": 0.5068,
+      "loss": 0.5154,
       "step": 920
     },
     {
       "epoch": 2.8725868725868726,
-      "grad_norm": 0.6735218011480151,
+      "grad_norm": 0.5919523831497919,
       "learning_rate": 5e-06,
-      "loss": 0.5076,
+      "loss": 0.516,
       "step": 930
     },
     {
       "epoch": 2.9034749034749034,
-      "grad_norm": 1.6394857976813213,
+      "grad_norm": 0.5425769518104601,
       "learning_rate": 5e-06,
-      "loss": 0.5032,
+      "loss": 0.5114,
       "step": 940
     },
     {
       "epoch": 2.9343629343629343,
-      "grad_norm": 0.5576873331550094,
+      "grad_norm": 0.49233460424864756,
       "learning_rate": 5e-06,
-      "loss": 0.504,
+      "loss": 0.5126,
       "step": 950
     },
     {
       "epoch": 2.965250965250965,
-      "grad_norm": 0.5637202117054834,
+      "grad_norm": 0.5105330314981293,
       "learning_rate": 5e-06,
-      "loss": 0.5012,
+      "loss": 0.5096,
       "step": 960
     },
     {
       "epoch": 2.993050193050193,
-      "eval_loss": 0.611685574054718,
-      "eval_runtime": 175.5767,
-      "eval_samples_per_second": 49.665,
+      "eval_loss": 0.6065478920936584,
+      "eval_runtime": 175.4708,
+      "eval_samples_per_second": 49.695,
       "eval_steps_per_second": 0.393,
       "step": 969
     },
@@ -708,9 +708,9 @@
       "epoch": 2.993050193050193,
       "step": 969,
       "total_flos": 1622692331520000.0,
-      "train_loss": 0.5660572312810719,
-      "train_runtime": 29124.935,
-      "train_samples_per_second": 17.065,
+      "train_loss": 0.5687421354839061,
+      "train_runtime": 29217.0698,
+      "train_samples_per_second": 17.012,
       "train_steps_per_second": 0.033
     }
   ],