Model save

README.md

@@ -1,11 +1,9 @@
 ---
 base_model: Qwen/Qwen2.5-1.5B-Instruct
-datasets: open-r1/OpenR1-Math-220k
 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 [open-r1/OpenR1-Math-220k](https://huggingface.co/datasets/open-r1/OpenR1-Math-220k) 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/wangguanghui0607/huggingface/runs/qxpmw0xc) 
+[<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/wangguanghui0607/huggingface/runs/3e0bojc6) 
 
 
 This model was trained with SFT.

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 488621249396736.0,
-    "train_loss": 0.0,
-    "train_runtime": 3.2409,
+    "train_loss": 0.5612765536157053,
+    "train_runtime": 11040.6024,
     "train_samples": 93733,
-    "train_samples_per_second": 10583.638,
-    "train_steps_per_second": 165.384
+    "train_samples_per_second": 3.107,
+    "train_steps_per_second": 0.049
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 488621249396736.0,
-    "train_loss": 0.0,
-    "train_runtime": 3.2409,
+    "train_loss": 0.5612765536157053,
+    "train_runtime": 11040.6024,
     "train_samples": 93733,
-    "train_samples_per_second": 10583.638,
-    "train_steps_per_second": 165.384
+    "train_samples_per_second": 3.107,
+    "train_steps_per_second": 0.049
 }

trainer_state.json

@@ -10,749 +10,749 @@
   "log_history": [
     {
       "epoch": 0.009328358208955223,
-      "grad_norm": 1.4534666584342113,
+      "grad_norm": 1.4539133207989317,
       "learning_rate": 9.259259259259259e-06,
       "loss": 0.8491,
       "step": 5
     },
     {
       "epoch": 0.018656716417910446,
-      "grad_norm": 1.1711916406560432,
+      "grad_norm": 1.1712962903753081,
       "learning_rate": 1.8518518518518518e-05,
       "loss": 0.8005,
       "step": 10
     },
     {
       "epoch": 0.027985074626865673,
-      "grad_norm": 0.7520916620233523,
+      "grad_norm": 0.7513244191485378,
       "learning_rate": 2.777777777777778e-05,
       "loss": 0.7313,
       "step": 15
     },
     {
       "epoch": 0.03731343283582089,
-      "grad_norm": 0.4675969780487494,
+      "grad_norm": 0.4664442595830348,
       "learning_rate": 3.7037037037037037e-05,
-      "loss": 0.6802,
+      "loss": 0.6803,
       "step": 20
     },
     {
       "epoch": 0.04664179104477612,
-      "grad_norm": 0.47274993852668085,
+      "grad_norm": 0.4730818672546804,
       "learning_rate": 4.62962962962963e-05,
       "loss": 0.6577,
       "step": 25
     },
     {
       "epoch": 0.055970149253731345,
-      "grad_norm": 0.4212378058530289,
+      "grad_norm": 0.42031591918358235,
       "learning_rate": 4.999614302517356e-05,
       "loss": 0.658,
       "step": 30
     },
     {
       "epoch": 0.06529850746268656,
-      "grad_norm": 0.3974247959215564,
+      "grad_norm": 0.399434425853996,
       "learning_rate": 4.997257741198456e-05,
       "loss": 0.6419,
       "step": 35
     },
     {
       "epoch": 0.07462686567164178,
-      "grad_norm": 0.3207977209065435,
+      "grad_norm": 0.31867989025493226,
       "learning_rate": 4.992761136351291e-05,
       "loss": 0.627,
       "step": 40
     },
     {
       "epoch": 0.08395522388059702,
-      "grad_norm": 0.34035380085701855,
+      "grad_norm": 0.3411350039428889,
       "learning_rate": 4.986128770052603e-05,
       "loss": 0.6084,
       "step": 45
     },
     {
       "epoch": 0.09328358208955224,
-      "grad_norm": 0.30354586362391395,
+      "grad_norm": 0.30161430505781994,
       "learning_rate": 4.9773669582457364e-05,
       "loss": 0.6101,
       "step": 50
     },
     {
       "epoch": 0.10261194029850747,
-      "grad_norm": 0.3866792254018339,
+      "grad_norm": 0.3824367863576047,
       "learning_rate": 4.966484044726024e-05,
       "loss": 0.6138,
       "step": 55
     },
     {
       "epoch": 0.11194029850746269,
-      "grad_norm": 0.37949830558516523,
+      "grad_norm": 0.37673227857223934,
       "learning_rate": 4.953490393195063e-05,
       "loss": 0.6006,
       "step": 60
     },
     {
       "epoch": 0.12126865671641791,
-      "grad_norm": 0.41179707666908344,
+      "grad_norm": 0.3940034789576713,
       "learning_rate": 4.938398377391461e-05,
-      "loss": 0.5975,
+      "loss": 0.5974,
       "step": 65
     },
     {
       "epoch": 0.13059701492537312,
-      "grad_norm": 0.5822016464409348,
+      "grad_norm": 0.5830770022640472,
       "learning_rate": 4.921222369307427e-05,
       "loss": 0.5941,
       "step": 70
     },
     {
       "epoch": 0.13992537313432835,
-      "grad_norm": 0.4786878847425219,
+      "grad_norm": 0.47445911407522817,
       "learning_rate": 4.901978725502454e-05,
       "loss": 0.5897,
       "step": 75
     },
     {
       "epoch": 0.14925373134328357,
-      "grad_norm": 0.4033723216228573,
+      "grad_norm": 0.41055806460856403,
       "learning_rate": 4.880685771527114e-05,
-      "loss": 0.5896,
+      "loss": 0.5895,
       "step": 80
     },
     {
       "epoch": 0.15858208955223882,
-      "grad_norm": 0.45477274659957395,
+      "grad_norm": 0.441633478829776,
       "learning_rate": 4.8573637844718e-05,
-      "loss": 0.5946,
+      "loss": 0.5945,
       "step": 85
     },
     {
       "epoch": 0.16791044776119404,
-      "grad_norm": 0.4591987379848635,
+      "grad_norm": 0.46583481627554624,
       "learning_rate": 4.83203497365703e-05,
-      "loss": 0.5913,
+      "loss": 0.5912,
       "step": 90
     },
     {
       "epoch": 0.17723880597014927,
-      "grad_norm": 0.512818276549736,
+      "grad_norm": 0.4902805881257132,
       "learning_rate": 4.8047234594837143e-05,
       "loss": 0.5783,
       "step": 95
     },
     {
       "epoch": 0.1865671641791045,
-      "grad_norm": 0.4685138256637361,
+      "grad_norm": 0.4647586357512984,
       "learning_rate": 4.775455250463507e-05,
-      "loss": 0.5751,
+      "loss": 0.575,
       "step": 100
     },
     {
       "epoch": 0.1958955223880597,
-      "grad_norm": 0.418117783574535,
+      "grad_norm": 0.39175526855873144,
       "learning_rate": 4.744258218451135e-05,
       "loss": 0.5849,
       "step": 105
     },
     {
       "epoch": 0.20522388059701493,
-      "grad_norm": 0.29569410470796165,
+      "grad_norm": 0.2982477617521291,
       "learning_rate": 4.71116207210228e-05,
       "loss": 0.5761,
       "step": 110
     },
     {
       "epoch": 0.21455223880597016,
-      "grad_norm": 0.36588086743641784,
+      "grad_norm": 0.35179378174248566,
       "learning_rate": 4.676198328582288e-05,
       "loss": 0.5713,
       "step": 115
     },
     {
       "epoch": 0.22388059701492538,
-      "grad_norm": 0.3341648966276976,
+      "grad_norm": 0.3305337106909977,
       "learning_rate": 4.6394002835526535e-05,
       "loss": 0.5822,
       "step": 120
     },
     {
       "epoch": 0.2332089552238806,
-      "grad_norm": 0.41521121469362104,
+      "grad_norm": 0.4162272011473445,
       "learning_rate": 4.6008029794638596e-05,
       "loss": 0.5668,
       "step": 125
     },
     {
       "epoch": 0.24253731343283583,
-      "grad_norm": 0.3961894491982948,
+      "grad_norm": 0.398913163115691,
       "learning_rate": 4.560443172184763e-05,
-      "loss": 0.5694,
+      "loss": 0.5695,
       "step": 130
     },
     {
       "epoch": 0.251865671641791,
-      "grad_norm": 0.3162166728580678,
+      "grad_norm": 0.31095345369274047,
       "learning_rate": 4.5183592960003104e-05,
       "loss": 0.562,
       "step": 135
     },
     {
       "epoch": 0.26119402985074625,
-      "grad_norm": 0.30927428386203404,
+      "grad_norm": 0.31460723326097384,
       "learning_rate": 4.4745914270109055e-05,
       "loss": 0.5673,
       "step": 140
     },
     {
       "epoch": 0.27052238805970147,
-      "grad_norm": 0.3536678799668726,
+      "grad_norm": 0.36633926737527456,
       "learning_rate": 4.429181244968301e-05,
       "loss": 0.5665,
       "step": 145
     },
     {
       "epoch": 0.2798507462686567,
-      "grad_norm": 0.33334109534004264,
+      "grad_norm": 0.33388276875216033,
       "learning_rate": 4.38217199358434e-05,
       "loss": 0.5605,
       "step": 150
     },
     {
       "epoch": 0.2891791044776119,
-      "grad_norm": 0.322653801766586,
+      "grad_norm": 0.3350871471924048,
       "learning_rate": 4.3336084393503545e-05,
       "loss": 0.5569,
       "step": 155
     },
     {
       "epoch": 0.29850746268656714,
-      "grad_norm": 0.3821614756667243,
+      "grad_norm": 0.3812826755487502,
       "learning_rate": 4.283536828906436e-05,
       "loss": 0.5693,
       "step": 160
     },
     {
       "epoch": 0.30783582089552236,
-      "grad_norm": 0.3833424325515942,
+      "grad_norm": 0.3866787149282935,
       "learning_rate": 4.2320048450011684e-05,
-      "loss": 0.5628,
+      "loss": 0.5627,
       "step": 165
     },
     {
       "epoch": 0.31716417910447764,
-      "grad_norm": 0.28204498731498767,
+      "grad_norm": 0.2839199189863909,
       "learning_rate": 4.179061561083777e-05,
       "loss": 0.5549,
       "step": 170
     },
     {
       "epoch": 0.32649253731343286,
-      "grad_norm": 0.4006876246661576,
+      "grad_norm": 0.3965758953952996,
       "learning_rate": 4.124757394571914e-05,
       "loss": 0.5613,
       "step": 175
     },
     {
       "epoch": 0.3358208955223881,
-      "grad_norm": 0.2991827926881174,
+      "grad_norm": 0.2953187179073866,
       "learning_rate": 4.069144058839605e-05,
       "loss": 0.5623,
       "step": 180
     },
     {
       "epoch": 0.3451492537313433,
-      "grad_norm": 0.36867658282010896,
+      "grad_norm": 0.36385487541980904,
       "learning_rate": 4.012274513971061e-05,
-      "loss": 0.5665,
+      "loss": 0.5664,
       "step": 185
     },
     {
       "epoch": 0.35447761194029853,
-      "grad_norm": 0.34656580201404646,
+      "grad_norm": 0.3485741950618337,
       "learning_rate": 3.954202916327264e-05,
       "loss": 0.5543,
       "step": 190
     },
     {
       "epoch": 0.36380597014925375,
-      "grad_norm": 0.32173574580213904,
+      "grad_norm": 0.3204032847588611,
       "learning_rate": 3.894984566973346e-05,
       "loss": 0.5545,
       "step": 195
     },
     {
       "epoch": 0.373134328358209,
-      "grad_norm": 0.29550755846319327,
+      "grad_norm": 0.29387286257840656,
       "learning_rate": 3.834675859015876e-05,
       "loss": 0.5584,
       "step": 200
     },
     {
       "epoch": 0.3824626865671642,
-      "grad_norm": 0.2981084523878612,
+      "grad_norm": 0.29955822569175505,
       "learning_rate": 3.77333422390021e-05,
       "loss": 0.5492,
       "step": 205
     },
     {
       "epoch": 0.3917910447761194,
-      "grad_norm": 0.3082421191846,
+      "grad_norm": 0.30176772096293986,
       "learning_rate": 3.711018076719034e-05,
-      "loss": 0.5564,
+      "loss": 0.5563,
       "step": 210
     },
     {
       "epoch": 0.40111940298507465,
-      "grad_norm": 0.3079273697879876,
+      "grad_norm": 0.3081564328345166,
       "learning_rate": 3.647786760584194e-05,
       "loss": 0.5607,
       "step": 215
     },
     {
       "epoch": 0.41044776119402987,
-      "grad_norm": 0.33213977610506984,
+      "grad_norm": 0.3170733557751109,
       "learning_rate": 3.583700490114776e-05,
       "loss": 0.5588,
       "step": 220
     },
     {
       "epoch": 0.4197761194029851,
-      "grad_norm": 0.3454312169959124,
+      "grad_norm": 0.34641631125524597,
       "learning_rate": 3.518820294095267e-05,
-      "loss": 0.5548,
+      "loss": 0.5547,
       "step": 225
     },
     {
       "epoch": 0.4291044776119403,
-      "grad_norm": 0.3111048651445072,
+      "grad_norm": 0.30393800144070166,
       "learning_rate": 3.453207957358377e-05,
       "loss": 0.5467,
       "step": 230
     },
     {
       "epoch": 0.43843283582089554,
-      "grad_norm": 0.3122159165573577,
+      "grad_norm": 0.31507435744899964,
       "learning_rate": 3.386925961947906e-05,
       "loss": 0.5478,
       "step": 235
     },
     {
       "epoch": 0.44776119402985076,
-      "grad_norm": 0.3815095659501264,
+      "grad_norm": 0.36211855059752174,
       "learning_rate": 3.320037427617639e-05,
       "loss": 0.5553,
       "step": 240
     },
     {
       "epoch": 0.457089552238806,
-      "grad_norm": 0.29471688390864853,
+      "grad_norm": 0.2944556611557611,
       "learning_rate": 3.252606051722972e-05,
       "loss": 0.5538,
       "step": 245
     },
     {
       "epoch": 0.4664179104477612,
-      "grad_norm": 0.3449459267022439,
+      "grad_norm": 0.35299495673417897,
       "learning_rate": 3.1846960485624886e-05,
       "loss": 0.5429,
       "step": 250
     },
     {
       "epoch": 0.47574626865671643,
-      "grad_norm": 0.3845030239986481,
+      "grad_norm": 0.3660273526350658,
       "learning_rate": 3.1163720882272516e-05,
       "loss": 0.5507,
       "step": 255
     },
     {
       "epoch": 0.48507462686567165,
-      "grad_norm": 0.3473121194428009,
+      "grad_norm": 0.338293029266179,
       "learning_rate": 3.047699235016056e-05,
-      "loss": 0.543,
+      "loss": 0.5429,
       "step": 260
     },
     {
       "epoch": 0.4944029850746269,
-      "grad_norm": 0.3183251881853174,
+      "grad_norm": 0.3254005319016007,
       "learning_rate": 2.9787428854752736e-05,
       "loss": 0.5368,
       "step": 265
     },
     {
       "epoch": 0.503731343283582,
-      "grad_norm": 0.28586873082292064,
+      "grad_norm": 0.2914208007253785,
       "learning_rate": 2.9095687061223058e-05,
       "loss": 0.5516,
       "step": 270
     },
     {
       "epoch": 0.5130597014925373,
-      "grad_norm": 0.34819444255911386,
+      "grad_norm": 0.3485366795929885,
       "learning_rate": 2.8402425709119435e-05,
       "loss": 0.5506,
       "step": 275
     },
     {
       "epoch": 0.5223880597014925,
-      "grad_norm": 0.2704108425263057,
+      "grad_norm": 0.27591964920917716,
       "learning_rate": 2.7708304985051868e-05,
       "loss": 0.5477,
       "step": 280
     },
     {
       "epoch": 0.5317164179104478,
-      "grad_norm": 0.28669112746143444,
+      "grad_norm": 0.2907217377958256,
       "learning_rate": 2.7013985894002623e-05,
-      "loss": 0.5356,
+      "loss": 0.5357,
       "step": 285
     },
     {
       "epoch": 0.5410447761194029,
-      "grad_norm": 0.2920252669556323,
+      "grad_norm": 0.30002064143622564,
       "learning_rate": 2.6320129629857093e-05,
       "loss": 0.5483,
       "step": 290
     },
     {
       "epoch": 0.5503731343283582,
-      "grad_norm": 0.2706020130328637,
+      "grad_norm": 0.27304160884463213,
       "learning_rate": 2.56273969457547e-05,
       "loss": 0.5373,
       "step": 295
     },
     {
       "epoch": 0.5597014925373134,
-      "grad_norm": 0.26211562115767434,
+      "grad_norm": 0.2701802509371835,
       "learning_rate": 2.4936447524859625e-05,
-      "loss": 0.5416,
+      "loss": 0.5417,
       "step": 300
     },
     {
       "epoch": 0.5690298507462687,
-      "grad_norm": 0.32486475734880277,
+      "grad_norm": 0.3307096329554463,
       "learning_rate": 2.4247939352150386e-05,
       "loss": 0.5367,
       "step": 305
     },
     {
       "epoch": 0.5783582089552238,
-      "grad_norm": 0.26365125864733874,
+      "grad_norm": 0.27363565184991906,
       "learning_rate": 2.3562528087826573e-05,
       "loss": 0.5429,
       "step": 310
     },
     {
       "epoch": 0.5876865671641791,
-      "grad_norm": 0.2513654842729793,
+      "grad_norm": 0.2461273976645459,
       "learning_rate": 2.2880866442929544e-05,
       "loss": 0.5398,
       "step": 315
     },
     {
       "epoch": 0.5970149253731343,
-      "grad_norm": 0.2429254461380733,
+      "grad_norm": 0.2374833064642032,
       "learning_rate": 2.2203603557771447e-05,
       "loss": 0.5359,
       "step": 320
     },
     {
       "epoch": 0.6063432835820896,
-      "grad_norm": 0.26495785362725305,
+      "grad_norm": 0.2648557827283419,
       "learning_rate": 2.153138438376473e-05,
       "loss": 0.5342,
       "step": 325
     },
     {
       "epoch": 0.6156716417910447,
-      "grad_norm": 0.24698377526769538,
+      "grad_norm": 0.24795560449882875,
       "learning_rate": 2.0864849069240645e-05,
       "loss": 0.5389,
       "step": 330
     },
     {
       "epoch": 0.625,
-      "grad_norm": 0.2716935501857811,
+      "grad_norm": 0.2765465631975035,
       "learning_rate": 2.0204632349841667e-05,
       "loss": 0.5357,
       "step": 335
     },
     {
       "epoch": 0.6343283582089553,
-      "grad_norm": 0.23513260099346342,
+      "grad_norm": 0.23594281950100068,
       "learning_rate": 1.9551362944068462e-05,
       "loss": 0.538,
       "step": 340
     },
     {
       "epoch": 0.6436567164179104,
-      "grad_norm": 0.26133157599374934,
+      "grad_norm": 0.25679843213592723,
       "learning_rate": 1.890566295455678e-05,
       "loss": 0.5313,
       "step": 345
     },
     {
       "epoch": 0.6529850746268657,
-      "grad_norm": 0.2449995496020128,
+      "grad_norm": 0.23816717672726997,
       "learning_rate": 1.8268147275654707e-05,
       "loss": 0.5411,
       "step": 350
     },
     {
       "epoch": 0.6623134328358209,
-      "grad_norm": 0.2247533847427069,
+      "grad_norm": 0.23250547728470577,
       "learning_rate": 1.7639423007864252e-05,
-      "loss": 0.5281,
+      "loss": 0.528,
       "step": 355
     },
     {
       "epoch": 0.6716417910447762,
-      "grad_norm": 0.2260024812533735,
+      "grad_norm": 0.22816436604168974,
       "learning_rate": 1.702008887970491e-05,
       "loss": 0.5289,
       "step": 360
     },
     {
       "epoch": 0.6809701492537313,
-      "grad_norm": 0.23618301620256973,
+      "grad_norm": 0.23417839793363304,
       "learning_rate": 1.6410734677549872e-05,
-      "loss": 0.5322,
+      "loss": 0.5321,
       "step": 365
     },
     {
       "epoch": 0.6902985074626866,
-      "grad_norm": 0.24235987409675933,
+      "grad_norm": 0.23779552425293224,
       "learning_rate": 1.58119406839777e-05,
       "loss": 0.5311,
       "step": 370
     },
     {
       "epoch": 0.6996268656716418,
-      "grad_norm": 0.2559397079642031,
+      "grad_norm": 0.256450902550975,
       "learning_rate": 1.5224277125174388e-05,
       "loss": 0.534,
       "step": 375
     },
     {
       "epoch": 0.7089552238805971,
-      "grad_norm": 0.2276944155439983,
+      "grad_norm": 0.2301008644434438,
       "learning_rate": 1.464830362791204e-05,
       "loss": 0.5363,
       "step": 380
     },
     {
       "epoch": 0.7182835820895522,
-      "grad_norm": 0.2199325902339066,
+      "grad_norm": 0.2228955428607329,
       "learning_rate": 1.4084568686621314e-05,
       "loss": 0.5386,
       "step": 385
     },
     {
       "epoch": 0.7276119402985075,
-      "grad_norm": 0.2188363369888737,
+      "grad_norm": 0.21981383409817995,
       "learning_rate": 1.3533609141065008e-05,
-      "loss": 0.5368,
+      "loss": 0.5369,
       "step": 390
     },
     {
       "epoch": 0.7369402985074627,
-      "grad_norm": 0.244089061405243,
+      "grad_norm": 0.2420445595904208,
       "learning_rate": 1.299594966511038e-05,
       "loss": 0.5341,
       "step": 395
     },
     {
       "epoch": 0.746268656716418,
-      "grad_norm": 0.22680536791217729,
+      "grad_norm": 0.2278233858059383,
       "learning_rate": 1.2472102267086904e-05,
       "loss": 0.5297,
       "step": 400
     },
     {
       "epoch": 0.7555970149253731,
-      "grad_norm": 0.2167560534054679,
+      "grad_norm": 0.21414740615604858,
       "learning_rate": 1.1962565802205255e-05,
       "loss": 0.5364,
       "step": 405
     },
     {
       "epoch": 0.7649253731343284,
-      "grad_norm": 0.2069416144688831,
+      "grad_norm": 0.20703259951041963,
       "learning_rate": 1.1467825497501954e-05,
       "loss": 0.5228,
       "step": 410
     },
     {
       "epoch": 0.7742537313432836,
-      "grad_norm": 0.2302676785561096,
+      "grad_norm": 0.2313972481386217,
       "learning_rate": 1.0988352489762006e-05,
       "loss": 0.5387,
       "step": 415
     },
     {
       "epoch": 0.7835820895522388,
-      "grad_norm": 0.2170000089714264,
+      "grad_norm": 0.21810415423257942,
       "learning_rate": 1.052460337685951e-05,
       "loss": 0.5301,
       "step": 420
     },
     {
       "epoch": 0.792910447761194,
-      "grad_norm": 0.21330280289188058,
+      "grad_norm": 0.21210174440030533,
       "learning_rate": 1.0077019782943584e-05,
       "loss": 0.5285,
       "step": 425
     },
     {
       "epoch": 0.8022388059701493,
-      "grad_norm": 0.22789671376553575,
+      "grad_norm": 0.226891037474942,
       "learning_rate": 9.646027937883622e-06,
       "loss": 0.5293,
       "step": 430
     },
     {
       "epoch": 0.8115671641791045,
-      "grad_norm": 0.2044388742976582,
+      "grad_norm": 0.20745817488065768,
       "learning_rate": 9.232038271374377e-06,
       "loss": 0.5312,
       "step": 435
     },
     {
       "epoch": 0.8208955223880597,
-      "grad_norm": 0.2121630276342737,
+      "grad_norm": 0.21040040659340928,
       "learning_rate": 8.835445022087426e-06,
-      "loss": 0.5259,
+      "loss": 0.5258,
       "step": 440
     },
     {
       "epoch": 0.8302238805970149,
-      "grad_norm": 0.21958440427518724,
+      "grad_norm": 0.2194474066244284,
       "learning_rate": 8.456625862241193e-06,
       "loss": 0.5361,
       "step": 445
     },
     {
       "epoch": 0.8395522388059702,
-      "grad_norm": 0.1993684141564352,
+      "grad_norm": 0.1969423352645484,
       "learning_rate": 8.095941537947057e-06,
       "loss": 0.5331,
       "step": 450
     },
     {
       "epoch": 0.8488805970149254,
-      "grad_norm": 0.22924886203423017,
+      "grad_norm": 0.2343265252202585,
       "learning_rate": 7.753735525674059e-06,
       "loss": 0.5258,
       "step": 455
     },
     {
       "epoch": 0.8582089552238806,
-      "grad_norm": 0.20514748374762057,
+      "grad_norm": 0.20537511829258023,
       "learning_rate": 7.430333705159286e-06,
       "loss": 0.5361,
       "step": 460
     },
     {
       "epoch": 0.8675373134328358,
-      "grad_norm": 0.19989921284599924,
+      "grad_norm": 0.20060828492971206,
       "learning_rate": 7.126044049075548e-06,
       "loss": 0.541,
       "step": 465
     },
     {
       "epoch": 0.8768656716417911,
-      "grad_norm": 0.19586385268397843,
+      "grad_norm": 0.19640210306856282,
       "learning_rate": 6.8411563297516995e-06,
       "loss": 0.5211,
       "step": 470
     },
     {
       "epoch": 0.8861940298507462,
-      "grad_norm": 0.1864657226226325,
+      "grad_norm": 0.18527466998099734,
       "learning_rate": 6.575941843225068e-06,
       "loss": 0.5249,
       "step": 475
     },
     {
       "epoch": 0.8955223880597015,
-      "grad_norm": 0.2190651926378328,
+      "grad_norm": 0.2153558358186585,
       "learning_rate": 6.330653150888617e-06,
-      "loss": 0.5294,
+      "loss": 0.5295,
       "step": 480
     },
     {
       "epoch": 0.9048507462686567,
-      "grad_norm": 0.2138242740322654,
+      "grad_norm": 0.21445098174513585,
       "learning_rate": 6.105523838979022e-06,
-      "loss": 0.5375,
+      "loss": 0.5376,
       "step": 485
     },
     {
       "epoch": 0.914179104477612,
-      "grad_norm": 0.19520032583669467,
+      "grad_norm": 0.19787448997224702,
       "learning_rate": 5.900768296134551e-06,
       "loss": 0.524,
       "step": 490
     },
     {
       "epoch": 0.9235074626865671,
-      "grad_norm": 0.1918945141121369,
+      "grad_norm": 0.19320876860048086,
       "learning_rate": 5.7165815092346825e-06,
       "loss": 0.5263,
       "step": 495
     },
     {
       "epoch": 0.9328358208955224,
-      "grad_norm": 0.18952383838163894,
+      "grad_norm": 0.19203704326975812,
       "learning_rate": 5.553138877715833e-06,
       "loss": 0.5281,
       "step": 500
     },
     {
       "epoch": 0.9421641791044776,
-      "grad_norm": 0.19849132534942884,
+      "grad_norm": 0.19792495046225278,
       "learning_rate": 5.410596046540051e-06,
       "loss": 0.5232,
       "step": 505
     },
     {
       "epoch": 0.9514925373134329,
-      "grad_norm": 0.21507884974780903,
+      "grad_norm": 0.21213741696029376,
       "learning_rate": 5.28908875797568e-06,
       "loss": 0.5237,
       "step": 510
     },
     {
       "epoch": 0.960820895522388,
-      "grad_norm": 0.19952690630617984,
+      "grad_norm": 0.1963524984996093,
       "learning_rate": 5.1887327223312296e-06,
       "loss": 0.5248,
       "step": 515
     },
     {
       "epoch": 0.9701492537313433,
-      "grad_norm": 0.2050907612918201,
+      "grad_norm": 0.20366670783823712,
       "learning_rate": 5.109623507765466e-06,
       "loss": 0.5291,
       "step": 520
     },
     {
       "epoch": 0.9794776119402985,
-      "grad_norm": 0.20534363644652603,
+      "grad_norm": 0.20440177097570653,
       "learning_rate": 5.051836449278715e-06,
       "loss": 0.5313,
       "step": 525
     },
     {
       "epoch": 0.9888059701492538,
-      "grad_norm": 0.20717020550482595,
+      "grad_norm": 0.20712434475203384,
       "learning_rate": 5.015426576972003e-06,
-      "loss": 0.5301,
+      "loss": 0.53,
       "step": 530
     },
     {
       "epoch": 0.9981343283582089,
-      "grad_norm": 0.190446884081656,
+      "grad_norm": 0.1914543360733239,
       "learning_rate": 5.000428563642382e-06,
       "loss": 0.5302,
       "step": 535
@@ -761,10 +761,10 @@
       "epoch": 1.0,
       "step": 536,
       "total_flos": 488621249396736.0,
-      "train_loss": 0.0,
-      "train_runtime": 3.2409,
-      "train_samples_per_second": 10583.638,
-      "train_steps_per_second": 165.384
+      "train_loss": 0.5612765536157053,
+      "train_runtime": 11040.6024,
+      "train_samples_per_second": 3.107,
+      "train_steps_per_second": 0.049
     }
   ],
   "logging_steps": 5,