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_DCFT_V3_wo_dataforge_economics
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # OH_DCFT_V3_wo_dataforge_economics
 
-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_DCFT_V3_wo_dataforge_economics dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.6378
 

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 2.9982238010657194,
-    "eval_loss": 0.6410338282585144,
-    "eval_runtime": 226.5313,
+    "eval_loss": 0.6377888917922974,
+    "eval_runtime": 226.5284,
     "eval_samples_per_second": 50.214,
     "eval_steps_per_second": 0.393,
     "total_flos": 2120178393415680.0,
-    "train_loss": 0.6170312019321026,
-    "train_runtime": 37849.9965,
-    "train_samples_per_second": 17.13,
+    "train_loss": 0.6200182046182159,
+    "train_runtime": 38034.9039,
+    "train_samples_per_second": 17.046,
     "train_steps_per_second": 0.033
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 2.9982238010657194,
-    "eval_loss": 0.6410338282585144,
-    "eval_runtime": 226.5313,
+    "eval_loss": 0.6377888917922974,
+    "eval_runtime": 226.5284,
     "eval_samples_per_second": 50.214,
     "eval_steps_per_second": 0.393
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 2.9982238010657194,
     "total_flos": 2120178393415680.0,
-    "train_loss": 0.6170312019321026,
-    "train_runtime": 37849.9965,
-    "train_samples_per_second": 17.13,
+    "train_loss": 0.6200182046182159,
+    "train_runtime": 38034.9039,
+    "train_samples_per_second": 17.046,
     "train_steps_per_second": 0.033
 }

trainer_state.json

@@ -10,917 +10,917 @@
   "log_history": [
     {
       "epoch": 0.023682652457075192,
-      "grad_norm": 1.816531801378568,
+      "grad_norm": 1.5009291573426993,
       "learning_rate": 5e-06,
-      "loss": 0.9186,
+      "loss": 0.8889,
       "step": 10
     },
     {
       "epoch": 0.047365304914150384,
-      "grad_norm": 1.3574653188634527,
+      "grad_norm": 1.2603811495508497,
       "learning_rate": 5e-06,
-      "loss": 0.8003,
+      "loss": 0.788,
       "step": 20
     },
     {
       "epoch": 0.07104795737122557,
-      "grad_norm": 1.5669400481998332,
+      "grad_norm": 0.8337069878140884,
       "learning_rate": 5e-06,
-      "loss": 0.7709,
+      "loss": 0.7543,
       "step": 30
     },
     {
       "epoch": 0.09473060982830077,
-      "grad_norm": 1.693017888571637,
+      "grad_norm": 0.7724138071764233,
       "learning_rate": 5e-06,
-      "loss": 0.7443,
+      "loss": 0.7321,
       "step": 40
     },
     {
       "epoch": 0.11841326228537596,
-      "grad_norm": 1.4722298325308145,
+      "grad_norm": 0.7604181173774881,
       "learning_rate": 5e-06,
-      "loss": 0.7337,
+      "loss": 0.7246,
       "step": 50
     },
     {
       "epoch": 0.14209591474245115,
-      "grad_norm": 0.6948406074694712,
+      "grad_norm": 0.7070103274073624,
       "learning_rate": 5e-06,
-      "loss": 0.7183,
+      "loss": 0.7112,
       "step": 60
     },
     {
       "epoch": 0.16577856719952636,
-      "grad_norm": 1.0658863949894195,
+      "grad_norm": 0.926548104967391,
       "learning_rate": 5e-06,
-      "loss": 0.7146,
+      "loss": 0.7088,
       "step": 70
     },
     {
       "epoch": 0.18946121965660154,
-      "grad_norm": 0.6741474876420499,
+      "grad_norm": 0.5558770280577042,
       "learning_rate": 5e-06,
-      "loss": 0.6984,
+      "loss": 0.6938,
       "step": 80
     },
     {
       "epoch": 0.21314387211367672,
-      "grad_norm": 1.0831273123966574,
+      "grad_norm": 0.6030257818397093,
       "learning_rate": 5e-06,
-      "loss": 0.7001,
+      "loss": 0.6949,
       "step": 90
     },
     {
       "epoch": 0.23682652457075193,
-      "grad_norm": 1.6635919094429064,
+      "grad_norm": 0.5336538650662063,
       "learning_rate": 5e-06,
-      "loss": 0.6886,
+      "loss": 0.6836,
       "step": 100
     },
     {
       "epoch": 0.2605091770278271,
-      "grad_norm": 0.8636474722702722,
+      "grad_norm": 0.8278552253316139,
       "learning_rate": 5e-06,
-      "loss": 0.686,
+      "loss": 0.6821,
       "step": 110
     },
     {
       "epoch": 0.2841918294849023,
-      "grad_norm": 0.7926351502503324,
+      "grad_norm": 0.6977545376176977,
       "learning_rate": 5e-06,
-      "loss": 0.6817,
+      "loss": 0.6782,
       "step": 120
     },
     {
       "epoch": 0.30787448194197753,
-      "grad_norm": 0.5901918000373718,
+      "grad_norm": 0.6610810719616435,
       "learning_rate": 5e-06,
-      "loss": 0.692,
+      "loss": 0.6889,
       "step": 130
     },
     {
       "epoch": 0.3315571343990527,
-      "grad_norm": 0.6148247600757576,
+      "grad_norm": 0.5558037806422206,
       "learning_rate": 5e-06,
-      "loss": 0.6817,
+      "loss": 0.6785,
       "step": 140
     },
     {
       "epoch": 0.3552397868561279,
-      "grad_norm": 0.5101477855796542,
+      "grad_norm": 0.636731427790056,
       "learning_rate": 5e-06,
-      "loss": 0.6853,
+      "loss": 0.6824,
       "step": 150
     },
     {
       "epoch": 0.3789224393132031,
-      "grad_norm": 0.5704569063807804,
+      "grad_norm": 0.5856745299156039,
       "learning_rate": 5e-06,
-      "loss": 0.6737,
+      "loss": 0.6713,
       "step": 160
     },
     {
       "epoch": 0.40260509177027826,
-      "grad_norm": 0.6027371888875237,
+      "grad_norm": 0.6346770000919194,
       "learning_rate": 5e-06,
-      "loss": 0.6774,
+      "loss": 0.6752,
       "step": 170
     },
     {
       "epoch": 0.42628774422735344,
-      "grad_norm": 0.6683781224223748,
+      "grad_norm": 0.5928544013121315,
       "learning_rate": 5e-06,
-      "loss": 0.6688,
+      "loss": 0.6667,
       "step": 180
     },
     {
       "epoch": 0.4499703966844287,
-      "grad_norm": 0.5016914580041181,
+      "grad_norm": 0.44291132727946475,
       "learning_rate": 5e-06,
-      "loss": 0.676,
+      "loss": 0.6741,
       "step": 190
     },
     {
       "epoch": 0.47365304914150386,
-      "grad_norm": 0.5452609890724252,
+      "grad_norm": 0.8183515848783448,
       "learning_rate": 5e-06,
-      "loss": 0.667,
+      "loss": 0.6657,
       "step": 200
     },
     {
       "epoch": 0.49733570159857904,
-      "grad_norm": 0.6510845954594989,
+      "grad_norm": 0.4527112724378749,
       "learning_rate": 5e-06,
-      "loss": 0.6691,
+      "loss": 0.6675,
       "step": 210
     },
     {
       "epoch": 0.5210183540556542,
-      "grad_norm": 0.5161096609174309,
+      "grad_norm": 0.4926733749500316,
       "learning_rate": 5e-06,
-      "loss": 0.6666,
+      "loss": 0.6652,
       "step": 220
     },
     {
       "epoch": 0.5447010065127295,
-      "grad_norm": 0.48148888185279093,
+      "grad_norm": 0.5595158055497421,
       "learning_rate": 5e-06,
-      "loss": 0.6705,
+      "loss": 0.6691,
       "step": 230
     },
     {
       "epoch": 0.5683836589698046,
-      "grad_norm": 0.646541503129204,
+      "grad_norm": 0.6125812031695264,
       "learning_rate": 5e-06,
-      "loss": 0.6615,
+      "loss": 0.6601,
       "step": 240
     },
     {
       "epoch": 0.5920663114268798,
-      "grad_norm": 0.5813160913392431,
+      "grad_norm": 0.48674298646689895,
       "learning_rate": 5e-06,
-      "loss": 0.6761,
+      "loss": 0.6747,
       "step": 250
     },
     {
       "epoch": 0.6157489638839551,
-      "grad_norm": 0.6228219376370357,
+      "grad_norm": 0.4613055270129497,
       "learning_rate": 5e-06,
-      "loss": 0.6645,
+      "loss": 0.6634,
       "step": 260
     },
     {
       "epoch": 0.6394316163410302,
-      "grad_norm": 0.532227126517761,
+      "grad_norm": 0.47597492722028856,
       "learning_rate": 5e-06,
-      "loss": 0.6634,
+      "loss": 0.6623,
       "step": 270
     },
     {
       "epoch": 0.6631142687981054,
-      "grad_norm": 0.5335777471459527,
+      "grad_norm": 0.4692966269330733,
       "learning_rate": 5e-06,
-      "loss": 0.6707,
+      "loss": 0.6695,
       "step": 280
     },
     {
       "epoch": 0.6867969212551805,
-      "grad_norm": 0.5859556398404715,
+      "grad_norm": 0.5086284371034546,
       "learning_rate": 5e-06,
-      "loss": 0.6595,
+      "loss": 0.6584,
       "step": 290
     },
     {
       "epoch": 0.7104795737122558,
-      "grad_norm": 0.678661754791658,
+      "grad_norm": 0.4936523688283777,
       "learning_rate": 5e-06,
-      "loss": 0.6635,
+      "loss": 0.6623,
       "step": 300
     },
     {
       "epoch": 0.7341622261693309,
-      "grad_norm": 0.5182311919791005,
+      "grad_norm": 0.5392570381158606,
       "learning_rate": 5e-06,
-      "loss": 0.6546,
+      "loss": 0.6536,
       "step": 310
     },
     {
       "epoch": 0.7578448786264061,
-      "grad_norm": 0.5694216979735989,
+      "grad_norm": 0.5646794511515978,
       "learning_rate": 5e-06,
-      "loss": 0.6541,
+      "loss": 0.6531,
       "step": 320
     },
     {
       "epoch": 0.7815275310834814,
-      "grad_norm": 0.5806274782090943,
+      "grad_norm": 0.4694116213218389,
       "learning_rate": 5e-06,
-      "loss": 0.6581,
+      "loss": 0.657,
       "step": 330
     },
     {
       "epoch": 0.8052101835405565,
-      "grad_norm": 0.57308874729709,
+      "grad_norm": 0.46701567257467963,
       "learning_rate": 5e-06,
-      "loss": 0.661,
+      "loss": 0.6601,
       "step": 340
     },
     {
       "epoch": 0.8288928359976317,
-      "grad_norm": 0.48566616576972127,
+      "grad_norm": 0.42590717154516355,
       "learning_rate": 5e-06,
-      "loss": 0.6514,
+      "loss": 0.6504,
       "step": 350
     },
     {
       "epoch": 0.8525754884547069,
-      "grad_norm": 0.5177674580432304,
+      "grad_norm": 0.45312976877971856,
       "learning_rate": 5e-06,
-      "loss": 0.6521,
+      "loss": 0.6512,
       "step": 360
     },
     {
       "epoch": 0.8762581409117821,
-      "grad_norm": 0.7662128776841642,
+      "grad_norm": 0.8468203203584982,
       "learning_rate": 5e-06,
-      "loss": 0.6521,
+      "loss": 0.6512,
       "step": 370
     },
     {
       "epoch": 0.8999407933688574,
-      "grad_norm": 0.7194268558213192,
+      "grad_norm": 0.542746765004271,
       "learning_rate": 5e-06,
-      "loss": 0.6527,
+      "loss": 0.652,
       "step": 380
     },
     {
       "epoch": 0.9236234458259325,
-      "grad_norm": 0.5317577993198188,
+      "grad_norm": 0.5833444889695804,
       "learning_rate": 5e-06,
-      "loss": 0.6462,
+      "loss": 0.6454,
       "step": 390
     },
     {
       "epoch": 0.9473060982830077,
-      "grad_norm": 0.6130071094405808,
+      "grad_norm": 0.6150273738613266,
       "learning_rate": 5e-06,
-      "loss": 0.6536,
+      "loss": 0.6529,
       "step": 400
     },
     {
       "epoch": 0.9709887507400828,
-      "grad_norm": 0.5054262813488729,
+      "grad_norm": 0.4798722057884747,
       "learning_rate": 5e-06,
-      "loss": 0.651,
+      "loss": 0.6503,
       "step": 410
     },
     {
       "epoch": 0.9946714031971581,
-      "grad_norm": 0.5216224953516784,
+      "grad_norm": 0.6351351683445448,
       "learning_rate": 5e-06,
-      "loss": 0.6504,
+      "loss": 0.6498,
       "step": 420
     },
     {
       "epoch": 0.9994079336885732,
-      "eval_loss": 0.6479789018630981,
-      "eval_runtime": 225.9605,
-      "eval_samples_per_second": 50.341,
-      "eval_steps_per_second": 0.394,
+      "eval_loss": 0.6475206017494202,
+      "eval_runtime": 225.2004,
+      "eval_samples_per_second": 50.511,
+      "eval_steps_per_second": 0.395,
       "step": 422
     },
     {
       "epoch": 1.0183540556542332,
-      "grad_norm": 0.6448097818743337,
+      "grad_norm": 0.5657496620777399,
       "learning_rate": 5e-06,
-      "loss": 0.6143,
+      "loss": 0.6177,
       "step": 430
     },
     {
       "epoch": 1.0420367081113084,
-      "grad_norm": 0.5557066787595923,
+      "grad_norm": 0.4892239120730945,
       "learning_rate": 5e-06,
-      "loss": 0.6112,
+      "loss": 0.6162,
       "step": 440
     },
     {
       "epoch": 1.0657193605683837,
-      "grad_norm": 0.6491926304674187,
+      "grad_norm": 0.5991744290901849,
       "learning_rate": 5e-06,
-      "loss": 0.5991,
+      "loss": 0.6042,
       "step": 450
     },
     {
       "epoch": 1.089402013025459,
-      "grad_norm": 0.5604079979316403,
+      "grad_norm": 0.44418054047007693,
       "learning_rate": 5e-06,
-      "loss": 0.6058,
+      "loss": 0.6107,
       "step": 460
     },
     {
       "epoch": 1.1130846654825342,
-      "grad_norm": 0.5772544741030392,
+      "grad_norm": 0.5035800519212378,
       "learning_rate": 5e-06,
-      "loss": 0.6075,
+      "loss": 0.6124,
       "step": 470
     },
     {
       "epoch": 1.1367673179396092,
-      "grad_norm": 0.4792859571457941,
+      "grad_norm": 0.4337632225925288,
       "learning_rate": 5e-06,
-      "loss": 0.6064,
+      "loss": 0.6113,
       "step": 480
     },
     {
       "epoch": 1.1604499703966844,
-      "grad_norm": 0.5655936542523597,
+      "grad_norm": 0.5274563247162665,
       "learning_rate": 5e-06,
-      "loss": 0.5994,
+      "loss": 0.6042,
       "step": 490
     },
     {
       "epoch": 1.1841326228537596,
-      "grad_norm": 0.7349705127084367,
+      "grad_norm": 0.6254329423737273,
       "learning_rate": 5e-06,
-      "loss": 0.6097,
+      "loss": 0.6144,
       "step": 500
     },
     {
       "epoch": 1.2078152753108349,
-      "grad_norm": 0.5763999652681121,
+      "grad_norm": 0.5562298009085542,
       "learning_rate": 5e-06,
-      "loss": 0.6135,
+      "loss": 0.6182,
       "step": 510
     },
     {
       "epoch": 1.2314979277679101,
-      "grad_norm": 0.6420586148204931,
+      "grad_norm": 0.6448107581853412,
       "learning_rate": 5e-06,
-      "loss": 0.6132,
+      "loss": 0.6178,
       "step": 520
     },
     {
       "epoch": 1.2551805802249851,
-      "grad_norm": 0.5788998804332586,
+      "grad_norm": 0.5597759186037163,
       "learning_rate": 5e-06,
-      "loss": 0.6004,
+      "loss": 0.6051,
       "step": 530
     },
     {
       "epoch": 1.2788632326820604,
-      "grad_norm": 0.536638979204567,
+      "grad_norm": 0.5094254785710688,
       "learning_rate": 5e-06,
-      "loss": 0.5964,
+      "loss": 0.601,
       "step": 540
     },
     {
       "epoch": 1.3025458851391356,
-      "grad_norm": 0.555896270898685,
+      "grad_norm": 0.4694431260792789,
       "learning_rate": 5e-06,
-      "loss": 0.6059,
+      "loss": 0.6106,
       "step": 550
     },
     {
       "epoch": 1.3262285375962108,
-      "grad_norm": 0.523881024195525,
+      "grad_norm": 0.47790257844004136,
       "learning_rate": 5e-06,
-      "loss": 0.6065,
+      "loss": 0.6109,
       "step": 560
     },
     {
       "epoch": 1.349911190053286,
-      "grad_norm": 0.5750577354690338,
+      "grad_norm": 0.5322942141679202,
       "learning_rate": 5e-06,
-      "loss": 0.6056,
+      "loss": 0.6101,
       "step": 570
     },
     {
       "epoch": 1.373593842510361,
-      "grad_norm": 0.5341219344965924,
+      "grad_norm": 0.45819277751159426,
       "learning_rate": 5e-06,
-      "loss": 0.6038,
+      "loss": 0.6081,
       "step": 580
     },
     {
       "epoch": 1.3972764949674363,
-      "grad_norm": 0.53243695293744,
+      "grad_norm": 0.4501719179556634,
       "learning_rate": 5e-06,
-      "loss": 0.6045,
+      "loss": 0.6089,
       "step": 590
     },
     {
       "epoch": 1.4209591474245116,
-      "grad_norm": 0.5160793462220815,
+      "grad_norm": 0.4542697399112238,
       "learning_rate": 5e-06,
-      "loss": 0.6109,
+      "loss": 0.6152,
       "step": 600
     },
     {
       "epoch": 1.4446417998815868,
-      "grad_norm": 0.519786480340386,
+      "grad_norm": 0.4553860474478459,
       "learning_rate": 5e-06,
-      "loss": 0.6047,
+      "loss": 0.6088,
       "step": 610
     },
     {
       "epoch": 1.468324452338662,
-      "grad_norm": 0.5176998387507717,
+      "grad_norm": 0.4652671956652514,
       "learning_rate": 5e-06,
-      "loss": 0.6025,
+      "loss": 0.6067,
       "step": 620
     },
     {
       "epoch": 1.492007104795737,
-      "grad_norm": 0.5326374678084742,
+      "grad_norm": 0.529442099891288,
       "learning_rate": 5e-06,
-      "loss": 0.6058,
+      "loss": 0.6102,
       "step": 630
     },
     {
       "epoch": 1.5156897572528123,
-      "grad_norm": 0.6814294915516854,
+      "grad_norm": 0.5263954654202675,
       "learning_rate": 5e-06,
-      "loss": 0.6018,
+      "loss": 0.606,
       "step": 640
     },
     {
       "epoch": 1.5393724097098875,
-      "grad_norm": 0.5067893932415504,
+      "grad_norm": 0.45995572644463967,
       "learning_rate": 5e-06,
-      "loss": 0.6093,
+      "loss": 0.6136,
       "step": 650
     },
     {
       "epoch": 1.5630550621669625,
-      "grad_norm": 0.6391993779464403,
+      "grad_norm": 0.5170488701848859,
       "learning_rate": 5e-06,
-      "loss": 0.6045,
+      "loss": 0.6086,
       "step": 660
     },
     {
       "epoch": 1.586737714624038,
-      "grad_norm": 0.5670108575671168,
+      "grad_norm": 0.4871128879860199,
       "learning_rate": 5e-06,
-      "loss": 0.6046,
+      "loss": 0.6086,
       "step": 670
     },
     {
       "epoch": 1.610420367081113,
-      "grad_norm": 0.5200973369062145,
+      "grad_norm": 0.4974631876345873,
       "learning_rate": 5e-06,
-      "loss": 0.6083,
+      "loss": 0.6125,
       "step": 680
     },
     {
       "epoch": 1.6341030195381883,
-      "grad_norm": 0.49629609919819867,
+      "grad_norm": 0.43324745659224545,
       "learning_rate": 5e-06,
-      "loss": 0.6102,
+      "loss": 0.6145,
       "step": 690
     },
     {
       "epoch": 1.6577856719952635,
-      "grad_norm": 0.5411312975043618,
+      "grad_norm": 0.5290498388557234,
       "learning_rate": 5e-06,
-      "loss": 0.6058,
+      "loss": 0.6099,
       "step": 700
     },
     {
       "epoch": 1.6814683244523385,
-      "grad_norm": 0.6511065530477933,
+      "grad_norm": 0.626734166124248,
       "learning_rate": 5e-06,
-      "loss": 0.6061,
+      "loss": 0.6101,
       "step": 710
     },
     {
       "epoch": 1.705150976909414,
-      "grad_norm": 0.5356040304164067,
+      "grad_norm": 0.5053972210175365,
       "learning_rate": 5e-06,
-      "loss": 0.608,
+      "loss": 0.6119,
       "step": 720
     },
     {
       "epoch": 1.728833629366489,
-      "grad_norm": 0.5044332661887174,
+      "grad_norm": 0.4918219678126516,
       "learning_rate": 5e-06,
-      "loss": 0.6067,
+      "loss": 0.6108,
       "step": 730
     },
     {
       "epoch": 1.7525162818235642,
-      "grad_norm": 0.5679268459481845,
+      "grad_norm": 0.47894871242204995,
       "learning_rate": 5e-06,
-      "loss": 0.6044,
+      "loss": 0.6083,
       "step": 740
     },
     {
       "epoch": 1.7761989342806395,
-      "grad_norm": 0.5041966525853055,
+      "grad_norm": 0.46473527001016457,
       "learning_rate": 5e-06,
-      "loss": 0.6012,
+      "loss": 0.6052,
       "step": 750
     },
     {
       "epoch": 1.7998815867377145,
-      "grad_norm": 0.622815173397146,
+      "grad_norm": 0.5249910352303092,
       "learning_rate": 5e-06,
-      "loss": 0.6064,
+      "loss": 0.6105,
       "step": 760
     },
     {
       "epoch": 1.82356423919479,
-      "grad_norm": 0.6105310434875696,
+      "grad_norm": 0.4939384702122645,
       "learning_rate": 5e-06,
-      "loss": 0.5993,
+      "loss": 0.6031,
       "step": 770
     },
     {
       "epoch": 1.847246891651865,
-      "grad_norm": 0.6478637621549127,
+      "grad_norm": 0.53408366438213,
       "learning_rate": 5e-06,
-      "loss": 0.6083,
+      "loss": 0.6121,
       "step": 780
     },
     {
       "epoch": 1.8709295441089402,
-      "grad_norm": 0.5288770221223392,
+      "grad_norm": 0.4462578146506802,
       "learning_rate": 5e-06,
-      "loss": 0.6064,
+      "loss": 0.6103,
       "step": 790
     },
     {
       "epoch": 1.8946121965660154,
-      "grad_norm": 0.6310494125165369,
+      "grad_norm": 0.5031337633152204,
       "learning_rate": 5e-06,
-      "loss": 0.6054,
+      "loss": 0.6092,
       "step": 800
     },
     {
       "epoch": 1.9182948490230904,
-      "grad_norm": 0.6224434949310766,
+      "grad_norm": 0.5118477000405415,
       "learning_rate": 5e-06,
-      "loss": 0.6014,
+      "loss": 0.6053,
       "step": 810
     },
     {
       "epoch": 1.941977501480166,
-      "grad_norm": 0.5690732220009218,
+      "grad_norm": 0.4778147349903897,
       "learning_rate": 5e-06,
-      "loss": 0.6026,
+      "loss": 0.6063,
       "step": 820
     },
     {
       "epoch": 1.965660153937241,
-      "grad_norm": 0.5819642741301115,
+      "grad_norm": 0.627775618514373,
       "learning_rate": 5e-06,
-      "loss": 0.6056,
+      "loss": 0.6096,
       "step": 830
     },
     {
       "epoch": 1.9893428063943162,
-      "grad_norm": 0.4867726663758881,
+      "grad_norm": 0.47474187825291425,
       "learning_rate": 5e-06,
-      "loss": 0.6039,
+      "loss": 0.6075,
       "step": 840
     },
     {
       "epoch": 1.9988158673771461,
-      "eval_loss": 0.6381043791770935,
-      "eval_runtime": 226.3666,
-      "eval_samples_per_second": 50.25,
-      "eval_steps_per_second": 0.393,
+      "eval_loss": 0.6369723081588745,
+      "eval_runtime": 225.3597,
+      "eval_samples_per_second": 50.475,
+      "eval_steps_per_second": 0.395,
       "step": 844
     },
     {
       "epoch": 2.0130254588513914,
-      "grad_norm": 0.799137975988829,
+      "grad_norm": 0.6033407582928859,
       "learning_rate": 5e-06,
-      "loss": 0.5789,
+      "loss": 0.5857,
       "step": 850
     },
     {
       "epoch": 2.0367081113084664,
-      "grad_norm": 0.5651415058987638,
+      "grad_norm": 0.48338661130504035,
       "learning_rate": 5e-06,
-      "loss": 0.5532,
+      "loss": 0.5626,
       "step": 860
     },
     {
       "epoch": 2.060390763765542,
-      "grad_norm": 0.6785997009798781,
+      "grad_norm": 0.5017850149423926,
       "learning_rate": 5e-06,
-      "loss": 0.5528,
+      "loss": 0.5622,
       "step": 870
     },
     {
       "epoch": 2.084073416222617,
-      "grad_norm": 0.589434711398729,
+      "grad_norm": 0.4965256354858608,
       "learning_rate": 5e-06,
-      "loss": 0.5581,
+      "loss": 0.5676,
       "step": 880
     },
     {
       "epoch": 2.1077560686796923,
-      "grad_norm": 0.6120240073948606,
+      "grad_norm": 0.5332089820071153,
       "learning_rate": 5e-06,
-      "loss": 0.5578,
+      "loss": 0.5673,
       "step": 890
     },
     {
       "epoch": 2.1314387211367674,
-      "grad_norm": 0.5356757525158238,
+      "grad_norm": 0.48166411012747123,
       "learning_rate": 5e-06,
-      "loss": 0.5564,
+      "loss": 0.5659,
       "step": 900
     },
     {
       "epoch": 2.1551213735938424,
-      "grad_norm": 0.6532444755899165,
+      "grad_norm": 0.4411082419748644,
       "learning_rate": 5e-06,
-      "loss": 0.5609,
+      "loss": 0.5702,
       "step": 910
     },
     {
       "epoch": 2.178804026050918,
-      "grad_norm": 0.7762170235905984,
+      "grad_norm": 0.561793122399257,
       "learning_rate": 5e-06,
-      "loss": 0.5588,
+      "loss": 0.5679,
       "step": 920
     },
     {
       "epoch": 2.202486678507993,
-      "grad_norm": 0.5110672320160398,
+      "grad_norm": 0.5150369649525713,
       "learning_rate": 5e-06,
-      "loss": 0.5611,
+      "loss": 0.5701,
       "step": 930
     },
     {
       "epoch": 2.2261693309650683,
-      "grad_norm": 0.5175930179811897,
+      "grad_norm": 0.5125519248465377,
       "learning_rate": 5e-06,
-      "loss": 0.5592,
+      "loss": 0.5682,
       "step": 940
     },
     {
       "epoch": 2.2498519834221433,
-      "grad_norm": 0.5781036208277457,
+      "grad_norm": 0.45758653804353955,
       "learning_rate": 5e-06,
-      "loss": 0.5567,
+      "loss": 0.5656,
       "step": 950
     },
     {
       "epoch": 2.2735346358792183,
-      "grad_norm": 0.638650838885228,
+      "grad_norm": 0.6337954299365447,
       "learning_rate": 5e-06,
-      "loss": 0.5566,
+      "loss": 0.5654,
       "step": 960
     },
     {
       "epoch": 2.297217288336294,
-      "grad_norm": 0.5928010322906232,
+      "grad_norm": 0.5455290125615978,
       "learning_rate": 5e-06,
-      "loss": 0.5595,
+      "loss": 0.5681,
       "step": 970
     },
     {
       "epoch": 2.320899940793369,
-      "grad_norm": 0.6471191913384154,
+      "grad_norm": 0.5503953364640523,
       "learning_rate": 5e-06,
-      "loss": 0.5618,
+      "loss": 0.5707,
       "step": 980
     },
     {
       "epoch": 2.3445825932504443,
-      "grad_norm": 0.5439480452105689,
+      "grad_norm": 0.47176571984059984,
       "learning_rate": 5e-06,
-      "loss": 0.5623,
+      "loss": 0.5709,
       "step": 990
     },
     {
       "epoch": 2.3682652457075193,
-      "grad_norm": 0.5616150107415377,
+      "grad_norm": 0.5002162272632372,
       "learning_rate": 5e-06,
-      "loss": 0.559,
+      "loss": 0.5677,
       "step": 1000
     },
     {
       "epoch": 2.3919478981645943,
-      "grad_norm": 0.6063862680614267,
+      "grad_norm": 0.5573828816208966,
       "learning_rate": 5e-06,
-      "loss": 0.5584,
+      "loss": 0.5666,
       "step": 1010
     },
     {
       "epoch": 2.4156305506216698,
-      "grad_norm": 0.6514652096821626,
+      "grad_norm": 0.5258827021378585,
       "learning_rate": 5e-06,
-      "loss": 0.564,
+      "loss": 0.5725,
       "step": 1020
     },
     {
       "epoch": 2.4393132030787448,
-      "grad_norm": 0.5978550068800899,
+      "grad_norm": 0.5589553231310385,
       "learning_rate": 5e-06,
-      "loss": 0.5648,
+      "loss": 0.5735,
       "step": 1030
     },
     {
       "epoch": 2.4629958555358202,
-      "grad_norm": 0.5389659551331341,
+      "grad_norm": 0.5100602577703085,
       "learning_rate": 5e-06,
-      "loss": 0.557,
+      "loss": 0.5653,
       "step": 1040
     },
     {
       "epoch": 2.4866785079928952,
-      "grad_norm": 0.49728611593057986,
+      "grad_norm": 0.4294738146845761,
       "learning_rate": 5e-06,
-      "loss": 0.5563,
+      "loss": 0.5646,
       "step": 1050
     },
     {
       "epoch": 2.5103611604499703,
-      "grad_norm": 0.5602610558768696,
+      "grad_norm": 0.548029454638703,
       "learning_rate": 5e-06,
-      "loss": 0.5628,
+      "loss": 0.5712,
       "step": 1060
     },
     {
       "epoch": 2.5340438129070457,
-      "grad_norm": 0.6603901062892502,
+      "grad_norm": 0.5514898162995253,
       "learning_rate": 5e-06,
-      "loss": 0.5576,
+      "loss": 0.5658,
       "step": 1070
     },
     {
       "epoch": 2.5577264653641207,
-      "grad_norm": 0.6506359439942624,
+      "grad_norm": 0.5518548821515326,
       "learning_rate": 5e-06,
-      "loss": 0.5567,
+      "loss": 0.5646,
       "step": 1080
     },
     {
       "epoch": 2.581409117821196,
-      "grad_norm": 0.6081261618456771,
+      "grad_norm": 0.5138811036011879,
       "learning_rate": 5e-06,
-      "loss": 0.5622,
+      "loss": 0.5703,
       "step": 1090
     },
     {
       "epoch": 2.605091770278271,
-      "grad_norm": 0.5583840013876924,
+      "grad_norm": 0.4749259332371036,
       "learning_rate": 5e-06,
-      "loss": 0.5641,
+      "loss": 0.5723,
       "step": 1100
     },
     {
       "epoch": 2.6287744227353462,
-      "grad_norm": 0.677285075920703,
+      "grad_norm": 0.6074130482912456,
       "learning_rate": 5e-06,
-      "loss": 0.5676,
+      "loss": 0.5759,
       "step": 1110
     },
     {
       "epoch": 2.6524570751924217,
-      "grad_norm": 0.5283556684105596,
+      "grad_norm": 0.4662137107433661,
       "learning_rate": 5e-06,
-      "loss": 0.5609,
+      "loss": 0.5693,
       "step": 1120
     },
     {
       "epoch": 2.6761397276494967,
-      "grad_norm": 0.5373459018769577,
+      "grad_norm": 0.47834715677436196,
       "learning_rate": 5e-06,
-      "loss": 0.5619,
+      "loss": 0.5701,
       "step": 1130
     },
     {
       "epoch": 2.699822380106572,
-      "grad_norm": 0.5544732747898201,
+      "grad_norm": 0.49830991624964954,
       "learning_rate": 5e-06,
-      "loss": 0.5577,
+      "loss": 0.5656,
       "step": 1140
     },
     {
       "epoch": 2.723505032563647,
-      "grad_norm": 0.5842636221616775,
+      "grad_norm": 0.6214961532916059,
       "learning_rate": 5e-06,
-      "loss": 0.5644,
+      "loss": 0.5722,
       "step": 1150
     },
     {
       "epoch": 2.747187685020722,
-      "grad_norm": 0.5195933914819354,
+      "grad_norm": 0.44581858116965695,
       "learning_rate": 5e-06,
-      "loss": 0.5669,
+      "loss": 0.5749,
       "step": 1160
     },
     {
       "epoch": 2.7708703374777977,
-      "grad_norm": 0.6191768570542974,
+      "grad_norm": 0.5288981698339921,
       "learning_rate": 5e-06,
-      "loss": 0.5647,
+      "loss": 0.5726,
       "step": 1170
     },
     {
       "epoch": 2.7945529899348727,
-      "grad_norm": 0.5155223116029976,
+      "grad_norm": 0.4343690821526064,
       "learning_rate": 5e-06,
-      "loss": 0.5681,
+      "loss": 0.5759,
       "step": 1180
     },
     {
       "epoch": 2.818235642391948,
-      "grad_norm": 0.6371568852312499,
+      "grad_norm": 0.5088546842925579,
       "learning_rate": 5e-06,
-      "loss": 0.5604,
+      "loss": 0.5681,
       "step": 1190
     },
     {
       "epoch": 2.841918294849023,
-      "grad_norm": 0.6170324836320265,
+      "grad_norm": 0.4739475280428968,
       "learning_rate": 5e-06,
-      "loss": 0.5634,
+      "loss": 0.5711,
       "step": 1200
     },
     {
       "epoch": 2.865600947306098,
-      "grad_norm": 0.5196345014766655,
+      "grad_norm": 0.4587392455090901,
       "learning_rate": 5e-06,
-      "loss": 0.5585,
+      "loss": 0.5662,
       "step": 1210
     },
     {
       "epoch": 2.8892835997631736,
-      "grad_norm": 0.5582532842634907,
+      "grad_norm": 0.5566523612983056,
       "learning_rate": 5e-06,
-      "loss": 0.5681,
+      "loss": 0.576,
       "step": 1220
     },
     {
       "epoch": 2.9129662522202486,
-      "grad_norm": 0.5370930409847251,
+      "grad_norm": 0.47828632727428366,
       "learning_rate": 5e-06,
-      "loss": 0.5622,
+      "loss": 0.5698,
       "step": 1230
     },
     {
       "epoch": 2.936648904677324,
-      "grad_norm": 0.5553114047738701,
+      "grad_norm": 0.4432815139860892,
       "learning_rate": 5e-06,
-      "loss": 0.5698,
+      "loss": 0.5776,
       "step": 1240
     },
     {
       "epoch": 2.960331557134399,
-      "grad_norm": 0.5109477407586265,
+      "grad_norm": 0.4472318974600453,
       "learning_rate": 5e-06,
-      "loss": 0.5651,
+      "loss": 0.5729,
       "step": 1250
     },
     {
       "epoch": 2.984014209591474,
-      "grad_norm": 0.535258137027712,
+      "grad_norm": 0.5067121626717436,
       "learning_rate": 5e-06,
-      "loss": 0.5552,
+      "loss": 0.5626,
       "step": 1260
     },
     {
       "epoch": 2.9982238010657194,
-      "eval_loss": 0.6410338282585144,
-      "eval_runtime": 226.7918,
-      "eval_samples_per_second": 50.156,
-      "eval_steps_per_second": 0.392,
+      "eval_loss": 0.6377888917922974,
+      "eval_runtime": 225.9066,
+      "eval_samples_per_second": 50.353,
+      "eval_steps_per_second": 0.394,
       "step": 1266
     },
     {
       "epoch": 2.9982238010657194,
       "step": 1266,
       "total_flos": 2120178393415680.0,
-      "train_loss": 0.6170312019321026,
-      "train_runtime": 37849.9965,
-      "train_samples_per_second": 17.13,
+      "train_loss": 0.6200182046182159,
+      "train_runtime": 38034.9039,
+      "train_samples_per_second": 17.046,
       "train_steps_per_second": 0.033
     }
   ],