Upload dpo_metrics.jsonl with huggingface_hub
Browse files- dpo_metrics.jsonl +73 -0
dpo_metrics.jsonl
CHANGED
|
@@ -217,3 +217,76 @@
|
|
| 217 |
{"timestamp": 1787715890.3798337, "elapsed_hours": 0.25063774678442213, "event": "dpo_step", "step": 2060, "loss": 0.7101656305789947, "pref_accuracy": 0.43, "chosen_reward": 0.12757262587547302, "rejected_reward": -0.16100655496120453, "lr": 1.731979951906069e-06, "tokens_per_sec": 151215.37925154835, "vram_gb": 4.253382682800293}
|
| 218 |
{"timestamp": 1787715893.4892375, "elapsed_hours": 0.25150147000948586, "event": "dpo_step", "step": 2070, "loss": 0.7098183488845825, "pref_accuracy": 0.48, "chosen_reward": -0.02989959716796875, "rejected_reward": -0.06517181545495987, "lr": 1.7299510609255612e-06, "tokens_per_sec": 151422.0531039978, "vram_gb": 3.790982723236084}
|
| 219 |
{"timestamp": 1787715897.4937878, "elapsed_hours": 0.2526138450702031, "event": "dpo_step", "step": 2080, "loss": 0.7083313012123108, "pref_accuracy": 0.46, "chosen_reward": -0.2057739496231079, "rejected_reward": -0.24403077363967896, "lr": 1.725874411861469e-06, "tokens_per_sec": 151476.48338967262, "vram_gb": 4.5285258293151855}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 217 |
{"timestamp": 1787715890.3798337, "elapsed_hours": 0.25063774678442213, "event": "dpo_step", "step": 2060, "loss": 0.7101656305789947, "pref_accuracy": 0.43, "chosen_reward": 0.12757262587547302, "rejected_reward": -0.16100655496120453, "lr": 1.731979951906069e-06, "tokens_per_sec": 151215.37925154835, "vram_gb": 4.253382682800293}
|
| 218 |
{"timestamp": 1787715893.4892375, "elapsed_hours": 0.25150147000948586, "event": "dpo_step", "step": 2070, "loss": 0.7098183488845825, "pref_accuracy": 0.48, "chosen_reward": -0.02989959716796875, "rejected_reward": -0.06517181545495987, "lr": 1.7299510609255612e-06, "tokens_per_sec": 151422.0531039978, "vram_gb": 3.790982723236084}
|
| 219 |
{"timestamp": 1787715897.4937878, "elapsed_hours": 0.2526138450702031, "event": "dpo_step", "step": 2080, "loss": 0.7083313012123108, "pref_accuracy": 0.46, "chosen_reward": -0.2057739496231079, "rejected_reward": -0.24403077363967896, "lr": 1.725874411861469e-06, "tokens_per_sec": 151476.48338967262, "vram_gb": 4.5285258293151855}
|
| 220 |
+
{"timestamp": 1787715902.185882, "elapsed_hours": 0.25391720453898114, "event": "dpo_step", "step": 2090, "loss": 0.7031788682937622, "pref_accuracy": 0.49, "chosen_reward": -0.029605865478515625, "rejected_reward": -0.00385284423828125, "lr": 1.7238266945204556e-06, "tokens_per_sec": 151415.25605587405, "vram_gb": 3.6844472885131836}
|
| 221 |
+
{"timestamp": 1787715905.2538369, "elapsed_hours": 0.2547694142659505, "event": "dpo_step", "step": 2100, "loss": 0.7002494645118713, "pref_accuracy": 0.54, "chosen_reward": -0.02610473707318306, "rejected_reward": -0.02038269117474556, "lr": 1.721772742572837e-06, "tokens_per_sec": 151625.49699544386, "vram_gb": 4.1898651123046875}
|
| 222 |
+
{"timestamp": 1787715909.8245292, "elapsed_hours": 0.25603905108239916, "event": "dpo_step", "step": 2110, "loss": 0.6961381018161774, "pref_accuracy": 0.56, "chosen_reward": -0.16071778535842896, "rejected_reward": 0.07010498642921448, "lr": 1.7197125765460808e-06, "tokens_per_sec": 151584.22489870366, "vram_gb": 4.956064224243164}
|
| 223 |
+
{"timestamp": 1787715913.375815, "elapsed_hours": 0.25702551934454176, "event": "dpo_step", "step": 2120, "loss": 0.6940844786167145, "pref_accuracy": 0.52, "chosen_reward": 0.05402832105755806, "rejected_reward": 0.10670928657054901, "lr": 1.7155736846753432e-06, "tokens_per_sec": 151712.00146659606, "vram_gb": 4.5285258293151855}
|
| 224 |
+
{"timestamp": 1787715917.8036344, "elapsed_hours": 0.25825546926922266, "event": "dpo_step", "step": 2130, "loss": 0.6841501903533935, "pref_accuracy": 0.57, "chosen_reward": -0.10345458984375, "rejected_reward": 0.03516846150159836, "lr": 1.713495000195995e-06, "tokens_per_sec": 151694.2039521243, "vram_gb": 4.956308364868164}
|
| 225 |
+
{"timestamp": 1787715920.5560043, "elapsed_hours": 0.25902001639207206, "event": "dpo_step", "step": 2140, "loss": 0.6879281830787659, "pref_accuracy": 0.55, "chosen_reward": -0.3229431211948395, "rejected_reward": 0.12284545600414276, "lr": 1.7114101843663633e-06, "tokens_per_sec": 151951.8780943496, "vram_gb": 4.942776203155518}
|
| 226 |
+
{"timestamp": 1787715923.6870294, "elapsed_hours": 0.2598897455135981, "event": "dpo_step", "step": 2150, "loss": 0.6921188199520111, "pref_accuracy": 0.53, "chosen_reward": 0.11467590928077698, "rejected_reward": 0.0148658761754632, "lr": 1.7093192580223732e-06, "tokens_per_sec": 152145.8042797985, "vram_gb": 4.94840669631958}
|
| 227 |
+
{"timestamp": 1787715927.1423862, "elapsed_hours": 0.26084956685702004, "event": "dpo_step", "step": 2160, "loss": 0.6938265347480774, "pref_accuracy": 0.52, "chosen_reward": 0.0243072509765625, "rejected_reward": -0.07563934475183487, "lr": 1.7051191574401578e-06, "tokens_per_sec": 152285.27152489423, "vram_gb": 4.5285258293151855}
|
| 228 |
+
{"timestamp": 1787715930.382729, "elapsed_hours": 0.2617496621608734, "event": "dpo_step", "step": 2170, "loss": 0.6957950043678284, "pref_accuracy": 0.55, "chosen_reward": 0.0559234619140625, "rejected_reward": -0.05685882642865181, "lr": 1.703010025178294e-06, "tokens_per_sec": 152458.841486477, "vram_gb": 3.965297222137451}
|
| 229 |
+
{"timestamp": 1787715935.1042178, "elapsed_hours": 0.2630611868699392, "event": "dpo_step", "step": 2180, "loss": 0.6969615733623504, "pref_accuracy": 0.51, "chosen_reward": 0.18644410371780396, "rejected_reward": -0.004357147961854935, "lr": 1.7008948663543754e-06, "tokens_per_sec": 152390.0724526627, "vram_gb": 4.668818473815918}
|
| 230 |
+
{"timestamp": 1787715939.4705691, "elapsed_hours": 0.2642740622493956, "event": "dpo_step", "step": 2190, "loss": 0.6957119202613831, "pref_accuracy": 0.49, "chosen_reward": -0.111846923828125, "rejected_reward": -0.18710099160671234, "lr": 1.6987737021075792e-06, "tokens_per_sec": 152379.42522977147, "vram_gb": 4.94943904876709}
|
| 231 |
+
{"timestamp": 1787715942.7842329, "elapsed_hours": 0.2651945243941413, "event": "dpo_step", "step": 2200, "loss": 0.6876695442199707, "pref_accuracy": 0.51, "chosen_reward": 0.06249389797449112, "rejected_reward": 0.08217544853687286, "lr": 1.6945134422019474e-06, "tokens_per_sec": 152538.5702555441, "vram_gb": 4.5285258293151855}
|
| 232 |
+
{"timestamp": 1787715945.9252267, "elapsed_hours": 0.2660670225487815, "event": "sample", "step": 2200, "sample_text": "Here is a Python function that returns the nth Fibonacci number:\n\n```python\ndef fibonacci(n):\n if n < 0:\n return 0\n elif n == 1:\n return 1\n elif n == 2:\n return 2\n else:\n return fibonacci(n-1) + fibonacci(n-2)\n\nn = int(input(\"Enter the number of Fibonacci numbers: \"))\nfibonacci(n)\n```\n"}
|
| 233 |
+
{"timestamp": 1787715950.239168, "elapsed_hours": 0.26726533969243366, "event": "dpo_step", "step": 2210, "loss": 0.689159461259842, "pref_accuracy": 0.48, "chosen_reward": 0.00391845777630806, "rejected_reward": -0.19925537705421448, "lr": 1.6923743891207135e-06, "tokens_per_sec": 152032.82342022678, "vram_gb": 3.8456978797912598}
|
| 234 |
+
{"timestamp": 1787715953.827925, "elapsed_hours": 0.26826221658123867, "event": "dpo_step", "step": 2220, "loss": 0.688459929227829, "pref_accuracy": 0.47, "chosen_reward": 0.06723709404468536, "rejected_reward": 0.005292129702866077, "lr": 1.69022941577138e-06, "tokens_per_sec": 152147.60348741236, "vram_gb": 4.9480133056640625}
|
| 235 |
+
{"timestamp": 1787715957.063062, "elapsed_hours": 0.269160865743955, "event": "dpo_step", "step": 2230, "loss": 0.7008815383911133, "pref_accuracy": 0.46, "chosen_reward": -0.10390014946460724, "rejected_reward": 0.0959014892578125, "lr": 1.688078543591095e-06, "tokens_per_sec": 152317.62458586553, "vram_gb": 4.101953983306885}
|
| 236 |
+
{"timestamp": 1787715961.4017503, "elapsed_hours": 0.27036605689260695, "event": "dpo_step", "step": 2240, "loss": 0.7022778415679931, "pref_accuracy": 0.45, "chosen_reward": -0.15140381455421448, "rejected_reward": 0.024462509900331497, "lr": 1.6837591887808137e-06, "tokens_per_sec": 152311.9217038066, "vram_gb": 4.29387903213501}
|
| 237 |
+
{"timestamp": 1787715965.6162922, "elapsed_hours": 0.2715367630455229, "event": "dpo_step", "step": 2250, "loss": 0.7125930309295654, "pref_accuracy": 0.43, "chosen_reward": -0.10218048095703125, "rejected_reward": -0.04033508151769638, "lr": 1.681590749319022e-06, "tokens_per_sec": 152325.80203692062, "vram_gb": 4.956613540649414}
|
| 238 |
+
{"timestamp": 1787715969.988721, "elapsed_hours": 0.2727513265609741, "event": "dpo_step", "step": 2260, "loss": 0.7133168530464172, "pref_accuracy": 0.46, "chosen_reward": -0.01717834547162056, "rejected_reward": 0.0022338852286338806, "lr": 1.679416497362255e-06, "tokens_per_sec": 152314.81880064085, "vram_gb": 4.016213893890381}
|
| 239 |
+
{"timestamp": 1787715972.9789064, "elapsed_hours": 0.27358193357785543, "event": "dpo_step", "step": 2270, "loss": 0.7199553871154785, "pref_accuracy": 0.45, "chosen_reward": -0.45380860567092896, "rejected_reward": 0.05353393405675888, "lr": 1.6772364546402749e-06, "tokens_per_sec": 152519.7739119984, "vram_gb": 4.943378925323486}
|
| 240 |
+
{"timestamp": 1787715976.391362, "elapsed_hours": 0.27452983803219266, "event": "dpo_step", "step": 2280, "loss": 0.7073396956920623, "pref_accuracy": 0.49, "chosen_reward": -0.17451781034469604, "rejected_reward": -0.15680542588233948, "lr": 1.6728590841088737e-06, "tokens_per_sec": 152657.59057997196, "vram_gb": 4.5285258293151855}
|
| 241 |
+
{"timestamp": 1787715981.2919486, "elapsed_hours": 0.27589111195670235, "event": "dpo_step", "step": 2290, "loss": 0.7134739398956299, "pref_accuracy": 0.53, "chosen_reward": 0.035868074744939804, "rejected_reward": -0.015302086248993874, "lr": 1.670661800047474e-06, "tokens_per_sec": 152563.3040250546, "vram_gb": 3.6893553733825684}
|
| 242 |
+
{"timestamp": 1787715984.9221995, "elapsed_hours": 0.27689951499303184, "event": "dpo_step", "step": 2300, "loss": 0.7155353713035584, "pref_accuracy": 0.51, "chosen_reward": 0.03803558275103569, "rejected_reward": 0.01187820453196764, "lr": 1.6684588127164654e-06, "tokens_per_sec": 152666.12733416475, "vram_gb": 3.9672722816467285}
|
| 243 |
+
{"timestamp": 1787715988.876431, "elapsed_hours": 0.27799791269832186, "event": "dpo_step", "step": 2310, "loss": 0.7240597665309906, "pref_accuracy": 0.46, "chosen_reward": 0.09378662705421448, "rejected_reward": 0.1065567135810852, "lr": 1.6662501441327953e-06, "tokens_per_sec": 152718.25646185823, "vram_gb": 4.955338954925537}
|
| 244 |
+
{"timestamp": 1787715994.0833948, "elapsed_hours": 0.2794442915254169, "event": "dpo_step", "step": 2320, "loss": 0.7209535002708435, "pref_accuracy": 0.44, "chosen_reward": 0.08469544351100922, "rejected_reward": 0.0600280798971653, "lr": 1.6618158515589343e-06, "tokens_per_sec": 152577.92678007408, "vram_gb": 4.5285258293151855}
|
| 245 |
+
{"timestamp": 1787715998.5189266, "elapsed_hours": 0.28067638370725845, "event": "dpo_step", "step": 2330, "loss": 0.7220729386806488, "pref_accuracy": 0.42, "chosen_reward": -0.03952636569738388, "rejected_reward": 0.09319000691175461, "lr": 1.6595902718856508e-06, "tokens_per_sec": 152556.54538447293, "vram_gb": 4.956064224243164}
|
| 246 |
+
{"timestamp": 1787716001.979851, "elapsed_hours": 0.28163775152630277, "event": "dpo_step", "step": 2340, "loss": 0.7276782596111298, "pref_accuracy": 0.41, "chosen_reward": 0.24659956991672516, "rejected_reward": 0.01054535061120987, "lr": 1.6573590995930776e-06, "tokens_per_sec": 152683.35736454936, "vram_gb": 4.506180286407471}
|
| 247 |
+
{"timestamp": 1787716006.1494179, "elapsed_hours": 0.28279596467812856, "event": "dpo_step", "step": 2350, "loss": 0.7195996689796448, "pref_accuracy": 0.44, "chosen_reward": 0.09927520900964737, "rejected_reward": 0.007823753170669079, "lr": 1.6551223569798458e-06, "tokens_per_sec": 152701.9395367404, "vram_gb": 4.956125259399414}
|
| 248 |
+
{"timestamp": 1787716008.0601532, "elapsed_hours": 0.283326724436548, "event": "dpo_step", "step": 2360, "loss": 0.720856466293335, "pref_accuracy": 0.47, "chosen_reward": -0.09079590439796448, "rejected_reward": 0.1932937651872635, "lr": 1.6506322502640615e-06, "tokens_per_sec": 153061.7538614589, "vram_gb": 3.524710178375244}
|
| 249 |
+
{"timestamp": 1787716011.8771627, "elapsed_hours": 0.28438700483904944, "event": "dpo_step", "step": 2370, "loss": 0.7202188324928284, "pref_accuracy": 0.47, "chosen_reward": -0.0257873572409153, "rejected_reward": -0.09279251098632812, "lr": 1.6483789310362298e-06, "tokens_per_sec": 153131.8747175642, "vram_gb": 4.956308364868164}
|
| 250 |
+
{"timestamp": 1787716015.6244013, "elapsed_hours": 0.28542790439393784, "event": "dpo_step", "step": 2380, "loss": 0.713777607679367, "pref_accuracy": 0.51, "chosen_reward": 0.23438109457492828, "rejected_reward": 0.17964211106300354, "lr": 1.646120131236734e-06, "tokens_per_sec": 153211.96787914293, "vram_gb": 4.945524215698242}
|
| 251 |
+
{"timestamp": 1787716020.625584, "elapsed_hours": 0.28681712177064683, "event": "dpo_step", "step": 2390, "loss": 0.7070296382904053, "pref_accuracy": 0.51, "chosen_reward": 0.10244293510913849, "rejected_reward": -0.09455566108226776, "lr": 1.6438558734403178e-06, "tokens_per_sec": 153103.58912508813, "vram_gb": 4.956186294555664}
|
| 252 |
+
{"timestamp": 1787716025.969818, "elapsed_hours": 0.2883016313446893, "event": "dpo_step", "step": 2400, "loss": 0.7134463214874267, "pref_accuracy": 0.48, "chosen_reward": 0.04158325493335724, "rejected_reward": 0.0144500732421875, "lr": 1.6393110744282155e-06, "tokens_per_sec": 152945.2179122591, "vram_gb": 4.5285258293151855}
|
| 253 |
+
{"timestamp": 1787716029.0837793, "elapsed_hours": 0.28916662057240805, "event": "sample", "step": 2400, "sample_text": "Here is a Python function that returns the nth Fibonacci number:\n\n```python\ndef fibonacci(n):\n if n <= 1:\n return n\n elif n == 2:\n return 1\n else:\n return fibonacci(n-1) + fibonacci(n-2)\n```\n\nThis function takes an integer `n` as input and returns the nth Fibonacci number. It first checks if `n` is equal to 1. If it is, it returns 1. If it is not, it"}
|
| 254 |
+
{"timestamp": 1787716032.527241, "elapsed_hours": 0.2901231375667784, "event": "dpo_step", "step": 2410, "loss": 0.7122899472713471, "pref_accuracy": 0.44, "chosen_reward": -0.18033447861671448, "rejected_reward": 0.08987732231616974, "lr": 1.6370305786338527e-06, "tokens_per_sec": 152609.3905342279, "vram_gb": 4.945285320281982}
|
| 255 |
+
{"timestamp": 1787716036.286346, "elapsed_hours": 0.29116733339097767, "event": "dpo_step", "step": 2420, "loss": 0.7062810242176056, "pref_accuracy": 0.48, "chosen_reward": -0.0638427734375, "rejected_reward": 0.01674804836511612, "lr": 1.6347447156847623e-06, "tokens_per_sec": 152688.0260733107, "vram_gb": 4.955085277557373}
|
| 256 |
+
{"timestamp": 1787716041.4665897, "elapsed_hours": 0.29260629011525047, "event": "dpo_step", "step": 2430, "loss": 0.7103239732980728, "pref_accuracy": 0.45, "chosen_reward": 0.03900451585650444, "rejected_reward": 0.008190155029296875, "lr": 1.6324535084261609e-06, "tokens_per_sec": 152558.12962958997, "vram_gb": 4.956308364868164}
|
| 257 |
+
{"timestamp": 1787716045.9246633, "elapsed_hours": 0.293844643831253, "event": "dpo_step", "step": 2440, "loss": 0.7109864860773086, "pref_accuracy": 0.43, "chosen_reward": -0.17730407416820526, "rejected_reward": 0.10699081420898438, "lr": 1.627855152628125e-06, "tokens_per_sec": 152534.51228882084, "vram_gb": 4.5285258293151855}
|
| 258 |
+
{"timestamp": 1787716050.1503365, "elapsed_hours": 0.2950184420082304, "event": "dpo_step", "step": 2450, "loss": 0.7036706799268723, "pref_accuracy": 0.48, "chosen_reward": -0.36839598417282104, "rejected_reward": -0.03316039964556694, "lr": 1.6255480500452667e-06, "tokens_per_sec": 152544.77610476792, "vram_gb": 4.956308364868164}
|
| 259 |
+
{"timestamp": 1787716053.9253647, "elapsed_hours": 0.29606706102689107, "event": "dpo_step", "step": 2460, "loss": 0.688072139620781, "pref_accuracy": 0.54, "chosen_reward": -0.04305725544691086, "rejected_reward": 0.16914673149585724, "lr": 1.623235695065592e-06, "tokens_per_sec": 152620.02980586307, "vram_gb": 4.956015110015869}
|
| 260 |
+
{"timestamp": 1787716058.0713546, "elapsed_hours": 0.2972187247541216, "event": "dpo_step", "step": 2470, "loss": 0.6804109579324722, "pref_accuracy": 0.55, "chosen_reward": -0.0044799805618822575, "rejected_reward": 0.05693969875574112, "lr": 1.6209181107990833e-06, "tokens_per_sec": 152641.3379434246, "vram_gb": 3.7461471557617188}
|
| 261 |
+
{"timestamp": 1787716063.1086555, "elapsed_hours": 0.2986179749833213, "event": "dpo_step", "step": 2480, "loss": 0.6715170180797577, "pref_accuracy": 0.58, "chosen_reward": 0.22138138115406036, "rejected_reward": -0.24342957139015198, "lr": 1.61626734710657e-06, "tokens_per_sec": 152534.77994418461, "vram_gb": 4.5285258293151855}
|
| 262 |
+
{"timestamp": 1787716066.985684, "elapsed_hours": 0.2996949273347855, "event": "dpo_step", "step": 2490, "loss": 0.6596467030048371, "pref_accuracy": 0.58, "chosen_reward": -0.06685180962085724, "rejected_reward": -0.010294342413544655, "lr": 1.6139342141609137e-06, "tokens_per_sec": 152594.61387482734, "vram_gb": 4.340306758880615}
|
| 263 |
+
{"timestamp": 1787716071.1252651, "elapsed_hours": 0.3008448110686408, "event": "dpo_step", "step": 2500, "loss": 0.652824729681015, "pref_accuracy": 0.58, "chosen_reward": 0.1894126832485199, "rejected_reward": -0.01510467566549778, "lr": 1.6115959448886551e-06, "tokens_per_sec": 152616.6715361905, "vram_gb": 4.974577903747559}
|
| 264 |
+
{"timestamp": 1787716081.0602922, "elapsed_hours": 0.30360454089111755, "event": "dpo_step", "step": 2510, "loss": 0.6642138671875, "pref_accuracy": 0.58, "chosen_reward": 0.10253067314624786, "rejected_reward": -0.05489044263958931, "lr": 1.6092525626587672e-06, "tokens_per_sec": 151822.1079049372, "vram_gb": 4.026832103729248}
|
| 265 |
+
{"timestamp": 1787716084.1655016, "elapsed_hours": 0.3044670989778307, "event": "dpo_step", "step": 2520, "loss": 0.672663289308548, "pref_accuracy": 0.57, "chosen_reward": -0.04714202880859375, "rejected_reward": 0.07558898627758026, "lr": 1.6045505530572575e-06, "tokens_per_sec": 151991.37286507268, "vram_gb": 4.5285258293151855}
|
| 266 |
+
{"timestamp": 1787716088.0672376, "elapsed_hours": 0.3055509145392312, "event": "dpo_step", "step": 2530, "loss": 0.6881666696071624, "pref_accuracy": 0.54, "chosen_reward": -0.4333862364292145, "rejected_reward": -0.13782958686351776, "lr": 1.6021919726781428e-06, "tokens_per_sec": 152048.53101554172, "vram_gb": 4.9532294273376465}
|
| 267 |
+
{"timestamp": 1787716092.0930207, "elapsed_hours": 0.3066691875457764, "event": "dpo_step", "step": 2540, "loss": 0.6823121780157089, "pref_accuracy": 0.59, "chosen_reward": 0.23129653930664062, "rejected_reward": -0.013671493157744408, "lr": 1.5998283733259423e-06, "tokens_per_sec": 152088.04978813117, "vram_gb": 4.271853923797607}
|
| 268 |
+
{"timestamp": 1787716096.421362, "elapsed_hours": 0.30787150455845724, "event": "dpo_step", "step": 2550, "loss": 0.690539944767952, "pref_accuracy": 0.56, "chosen_reward": -0.294952392578125, "rejected_reward": -0.17734679579734802, "lr": 1.597459778622781e-06, "tokens_per_sec": 152085.37940714718, "vram_gb": 4.956064224243164}
|
| 269 |
+
{"timestamp": 1787716101.0504878, "elapsed_hours": 0.30915737284554373, "event": "dpo_step", "step": 2560, "loss": 0.6963602870702743, "pref_accuracy": 0.55, "chosen_reward": 0.25577086210250854, "rejected_reward": 0.1410781890153885, "lr": 1.5927076979014604e-06, "tokens_per_sec": 152041.2749652985, "vram_gb": 4.5285258293151855}
|
| 270 |
+
{"timestamp": 1787716106.0728412, "elapsed_hours": 0.3105524710814158, "event": "dpo_step", "step": 2570, "loss": 0.6885219073295593, "pref_accuracy": 0.56, "chosen_reward": 0.08599244058132172, "rejected_reward": 0.17887572944164276, "lr": 1.5903242593762252e-06, "tokens_per_sec": 151943.62539017512, "vram_gb": 4.956125259399414}
|
| 271 |
+
{"timestamp": 1787716111.0103862, "elapsed_hours": 0.31192401130994163, "event": "dpo_step", "step": 2580, "loss": 0.671825270652771, "pref_accuracy": 0.59, "chosen_reward": 0.07515869289636612, "rejected_reward": 0.01510619930922985, "lr": 1.5879359204854028e-06, "tokens_per_sec": 151858.4184074617, "vram_gb": 4.956247329711914}
|
| 272 |
+
{"timestamp": 1787716115.3238943, "elapsed_hours": 0.31312220798598395, "event": "dpo_step", "step": 2590, "loss": 0.700936262011528, "pref_accuracy": 0.52, "chosen_reward": 0.06473693996667862, "rejected_reward": 0.04624138027429581, "lr": 1.5855427050983682e-06, "tokens_per_sec": 151858.7054364051, "vram_gb": 3.8426613807678223}
|
| 273 |
+
{"timestamp": 1787716119.2016053, "elapsed_hours": 0.3141993500126733, "event": "dpo_step", "step": 2600, "loss": 0.7046817463636398, "pref_accuracy": 0.49, "chosen_reward": -0.08753661811351776, "rejected_reward": -0.14203643798828125, "lr": 1.5807417405566067e-06, "tokens_per_sec": 151917.99008159418, "vram_gb": 3.7488622665405273}
|
| 274 |
+
{"timestamp": 1787716122.4457917, "elapsed_hours": 0.31510051283571455, "event": "sample", "step": 2600, "sample_text": "Here is a Python function that takes a number as input and returns the nth Fibonacci number:\n\n```python\ndef fibonacci(n):\n if n <= 0:\n return 1\n elif n == 1:\n return n * 2\n else:\n return fibonacci(n - 1) + fibonacci(n - 2)\n\nn = 10\nprint(fibonacci(n))\n```\n\nWhen you run this code, it will output: `1 2 3 4 5 6 7 8 9 10 11 12"}
|
| 275 |
+
{"timestamp": 1787716125.8035567, "elapsed_hours": 0.3160332253244188, "event": "dpo_step", "step": 2610, "loss": 0.6967265492677689, "pref_accuracy": 0.49, "chosen_reward": -0.04360351711511612, "rejected_reward": -0.13081054389476776, "lr": 1.5783340393833552e-06, "tokens_per_sec": 151609.9043034121, "vram_gb": 4.943459987640381}
|
| 276 |
+
{"timestamp": 1787716130.4008298, "elapsed_hours": 0.3173102456331253, "event": "dpo_step", "step": 2620, "loss": 0.7124523508548737, "pref_accuracy": 0.47, "chosen_reward": 0.12846679985523224, "rejected_reward": 0.34648436307907104, "lr": 1.5759215576763626e-06, "tokens_per_sec": 151573.14521349414, "vram_gb": 4.956308364868164}
|
| 277 |
+
{"timestamp": 1787716134.4127364, "elapsed_hours": 0.3184246640735202, "event": "dpo_step", "step": 2630, "loss": 0.7323345637321472, "pref_accuracy": 0.45, "chosen_reward": -0.5458862781524658, "rejected_reward": 0.27867433428764343, "lr": 1.5735043195462917e-06, "tokens_per_sec": 151614.71924819273, "vram_gb": 4.956104278564453}
|
| 278 |
+
{"timestamp": 1787716138.32318, "elapsed_hours": 0.3195108985900879, "event": "dpo_step", "step": 2640, "loss": 0.7198031723499299, "pref_accuracy": 0.49, "chosen_reward": 0.25649717450141907, "rejected_reward": 0.007000732701271772, "lr": 1.5686556706969993e-06, "tokens_per_sec": 151669.4910930298, "vram_gb": 4.5285258293151855}
|
| 279 |
+
{"timestamp": 1787716142.0535169, "elapsed_hours": 0.32054710315333473, "event": "dpo_step", "step": 2650, "loss": 0.6995050221681595, "pref_accuracy": 0.57, "chosen_reward": -0.1103363111615181, "rejected_reward": -0.01753539964556694, "lr": 1.566224308435817e-06, "tokens_per_sec": 151747.76045039773, "vram_gb": 3.8751349449157715}
|
| 280 |
+
{"timestamp": 1787716147.096731, "elapsed_hours": 0.3219479960203171, "event": "dpo_step", "step": 2660, "loss": 0.6955326384305954, "pref_accuracy": 0.56, "chosen_reward": 0.02244873158633709, "rejected_reward": 0.20013427734375, "lr": 1.5637882866671492e-06, "tokens_per_sec": 151652.12833724622, "vram_gb": 4.956308364868164}
|
| 281 |
+
{"timestamp": 1787716152.0168676, "elapsed_hours": 0.32331470059023965, "event": "dpo_step", "step": 2670, "loss": 0.6864810079336167, "pref_accuracy": 0.6, "chosen_reward": 0.2729248106479645, "rejected_reward": -0.044677734375, "lr": 1.5613476297369198e-06, "tokens_per_sec": 151573.48450302923, "vram_gb": 4.956308364868164}
|
| 282 |
+
{"timestamp": 1787716156.5744731, "elapsed_hours": 0.32458070211940343, "event": "dpo_step", "step": 2680, "loss": 0.6775338667631149, "pref_accuracy": 0.59, "chosen_reward": -0.16628876328468323, "rejected_reward": -0.1940658688545227, "lr": 1.5564525080068512e-06, "tokens_per_sec": 151542.8979732832, "vram_gb": 4.5285258293151855}
|
| 283 |
+
{"timestamp": 1787716161.093657, "elapsed_hours": 0.32583603110578324, "event": "dpo_step", "step": 2690, "loss": 0.6777534586191177, "pref_accuracy": 0.57, "chosen_reward": -0.007537841796875, "rejected_reward": 0.1702495515346527, "lr": 1.5539980921295075e-06, "tokens_per_sec": 151517.54856620042, "vram_gb": 4.956247329711914}
|
| 284 |
+
{"timestamp": 1787716164.3083162, "elapsed_hours": 0.32672899186611176, "event": "dpo_step", "step": 2700, "loss": 0.6915747642517089, "pref_accuracy": 0.53, "chosen_reward": 0.12105255573987961, "rejected_reward": 0.0617218017578125, "lr": 1.5515391389351025e-06, "tokens_per_sec": 151661.78332290868, "vram_gb": 4.118798732757568}
|
| 285 |
+
{"timestamp": 1787716169.2019036, "elapsed_hours": 0.328088321685791, "event": "dpo_step", "step": 2710, "loss": 0.7030907535552978, "pref_accuracy": 0.53, "chosen_reward": 0.0007232669740915298, "rejected_reward": -0.23128661513328552, "lr": 1.549075672998741e-06, "tokens_per_sec": 151587.68842868618, "vram_gb": 4.956186294555664}
|
| 286 |
+
{"timestamp": 1787716173.8324387, "elapsed_hours": 0.3293745814429389, "event": "dpo_step", "step": 2720, "loss": 0.6953129780292511, "pref_accuracy": 0.51, "chosen_reward": 0.34306642413139343, "rejected_reward": -0.160888671875, "lr": 1.5441353014258217e-06, "tokens_per_sec": 151548.09357576762, "vram_gb": 4.5285258293151855}
|
| 287 |
+
{"timestamp": 1787716177.6786575, "elapsed_hours": 0.3304429755608241, "event": "dpo_step", "step": 2730, "loss": 0.6980029487609863, "pref_accuracy": 0.5, "chosen_reward": -0.09034577012062073, "rejected_reward": -0.05872192233800888, "lr": 1.5416584451639919e-06, "tokens_per_sec": 151609.50992642256, "vram_gb": 4.95512056350708}
|
| 288 |
+
{"timestamp": 1787716181.4107401, "elapsed_hours": 0.33147966530587936, "event": "dpo_step", "step": 2740, "loss": 0.7044875657558441, "pref_accuracy": 0.49, "chosen_reward": -0.1886184811592102, "rejected_reward": 0.06684722751379013, "lr": 1.5391771749091673e-06, "tokens_per_sec": 151685.14662845054, "vram_gb": 4.974752902984619}
|
| 289 |
+
{"timestamp": 1787716186.422765, "elapsed_hours": 0.33287189430660674, "event": "dpo_step", "step": 2750, "loss": 0.7103543019294739, "pref_accuracy": 0.48, "chosen_reward": -0.04276123270392418, "rejected_reward": 0.08344727009534836, "lr": 1.5366915154594915e-06, "tokens_per_sec": 151596.9167263459, "vram_gb": 4.956308364868164}
|
| 290 |
+
{"timestamp": 1787716190.460649, "elapsed_hours": 0.3339935287502077, "event": "dpo_step", "step": 2760, "loss": 0.709025399684906, "pref_accuracy": 0.48, "chosen_reward": -0.1113433837890625, "rejected_reward": 0.1878509521484375, "lr": 1.531707128387244e-06, "tokens_per_sec": 151633.15775519225, "vram_gb": 4.5285258293151855}
|
| 291 |
+
{"timestamp": 1787716195.1347816, "elapsed_hours": 0.3352918988466263, "event": "dpo_step", "step": 2770, "loss": 0.723406343460083, "pref_accuracy": 0.47, "chosen_reward": 0.11997070908546448, "rejected_reward": 0.18498899042606354, "lr": 1.5292084505792967e-06, "tokens_per_sec": 151588.56912585793, "vram_gb": 4.592715740203857}
|
| 292 |
+
{"timestamp": 1787716199.3872063, "elapsed_hours": 0.33647312813334995, "event": "dpo_step", "step": 2780, "loss": 0.7145835375785827, "pref_accuracy": 0.52, "chosen_reward": -0.01452636905014515, "rejected_reward": -0.5156311392784119, "lr": 1.5267054832052494e-06, "tokens_per_sec": 151597.50878709852, "vram_gb": 4.956247329711914}
|