Upload dpo_metrics.jsonl with huggingface_hub
Browse files- dpo_metrics.jsonl +71 -0
dpo_metrics.jsonl
CHANGED
|
@@ -146,3 +146,74 @@
|
|
| 146 |
{"timestamp": 1787715589.4235928, "elapsed_hours": 0.16703879098097482, "event": "dpo_step", "step": 1390, "loss": 0.6596014487743378, "pref_accuracy": 0.63, "chosen_reward": 0.010083010420203209, "rejected_reward": 0.14625243842601776, "lr": 1.877996362953877e-06, "tokens_per_sec": 153918.9682018877, "vram_gb": 4.956308364868164}
|
| 147 |
{"timestamp": 1787715592.8423104, "elapsed_hours": 0.1679884346988466, "event": "dpo_step", "step": 1400, "loss": 0.6682743489742279, "pref_accuracy": 0.63, "chosen_reward": 0.06715889275074005, "rejected_reward": 0.03865242004394531, "lr": 1.8751200501879146e-06, "tokens_per_sec": 154135.93025515217, "vram_gb": 3.751861095428467}
|
| 148 |
{"timestamp": 1787715596.0137901, "elapsed_hours": 0.1688694014151891, "event": "sample", "step": 1400, "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 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 uses a recursive approach to find the nth Fibonacci number.\n\nThe function iterates through each number `n` in the range from 2 to the square root of `n`. The `fibonacci"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 146 |
{"timestamp": 1787715589.4235928, "elapsed_hours": 0.16703879098097482, "event": "dpo_step", "step": 1390, "loss": 0.6596014487743378, "pref_accuracy": 0.63, "chosen_reward": 0.010083010420203209, "rejected_reward": 0.14625243842601776, "lr": 1.877996362953877e-06, "tokens_per_sec": 153918.9682018877, "vram_gb": 4.956308364868164}
|
| 147 |
{"timestamp": 1787715592.8423104, "elapsed_hours": 0.1679884346988466, "event": "dpo_step", "step": 1400, "loss": 0.6682743489742279, "pref_accuracy": 0.63, "chosen_reward": 0.06715889275074005, "rejected_reward": 0.03865242004394531, "lr": 1.8751200501879146e-06, "tokens_per_sec": 154135.93025515217, "vram_gb": 3.751861095428467}
|
| 148 |
{"timestamp": 1787715596.0137901, "elapsed_hours": 0.1688694014151891, "event": "sample", "step": 1400, "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 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 uses a recursive approach to find the nth Fibonacci number.\n\nThe function iterates through each number `n` in the range from 2 to the square root of `n`. The `fibonacci"}
|
| 149 |
+
{"timestamp": 1787715600.3223286, "elapsed_hours": 0.17006621758143106, "event": "dpo_step", "step": 1410, "loss": 0.6638955855369568, "pref_accuracy": 0.63, "chosen_reward": 0.26948243379592896, "rejected_reward": -0.29283449053764343, "lr": 1.8736702666203206e-06, "tokens_per_sec": 153310.4056612962, "vram_gb": 4.955927848815918}
|
| 150 |
+
{"timestamp": 1787715603.8235862, "elapsed_hours": 0.1710387890206443, "event": "dpo_step", "step": 1420, "loss": 0.6827209210395813, "pref_accuracy": 0.56, "chosen_reward": -0.01020202785730362, "rejected_reward": -0.004509735852479935, "lr": 1.8722127508896166e-06, "tokens_per_sec": 153506.00123687586, "vram_gb": 4.083016395568848}
|
| 151 |
+
{"timestamp": 1787715607.118646, "elapsed_hours": 0.17195408357514275, "event": "dpo_step", "step": 1430, "loss": 0.6940959429740906, "pref_accuracy": 0.52, "chosen_reward": -0.04013671725988388, "rejected_reward": 0.13576659560203552, "lr": 1.8707475175624075e-06, "tokens_per_sec": 153751.3573165825, "vram_gb": 4.940478801727295}
|
| 152 |
+
{"timestamp": 1787715612.2135968, "elapsed_hours": 0.17336934778425428, "event": "dpo_step", "step": 1440, "loss": 0.7029692447185516, "pref_accuracy": 0.5, "chosen_reward": 0.17371827363967896, "rejected_reward": 0.0245361328125, "lr": 1.8677939567704021e-06, "tokens_per_sec": 153543.10309445343, "vram_gb": 4.5285258293151855}
|
| 153 |
+
{"timestamp": 1787715616.2540808, "elapsed_hours": 0.174491704305013, "event": "dpo_step", "step": 1450, "loss": 0.7068187594413757, "pref_accuracy": 0.46, "chosen_reward": -0.11670227348804474, "rejected_reward": -0.07125091552734375, "lr": 1.8663056588238835e-06, "tokens_per_sec": 153599.63221144772, "vram_gb": 4.954769134521484}
|
| 154 |
+
{"timestamp": 1787715620.5961592, "elapsed_hours": 0.17569783720705245, "event": "dpo_step", "step": 1460, "loss": 0.7018823724985123, "pref_accuracy": 0.5, "chosen_reward": 0.03448868170380592, "rejected_reward": -0.10774984210729599, "lr": 1.8648097023171208e-06, "tokens_per_sec": 153581.04197268668, "vram_gb": 4.333329200744629}
|
| 155 |
+
{"timestamp": 1787715624.7076454, "elapsed_hours": 0.17683991663985782, "event": "dpo_step", "step": 1470, "loss": 0.6949162524938584, "pref_accuracy": 0.53, "chosen_reward": 0.3774704039096832, "rejected_reward": -0.07317047566175461, "lr": 1.8633061022009007e-06, "tokens_per_sec": 153619.16924890238, "vram_gb": 4.936587810516357}
|
| 156 |
+
{"timestamp": 1787715627.9427636, "elapsed_hours": 0.17773856063683827, "event": "dpo_step", "step": 1480, "loss": 0.6897094053030014, "pref_accuracy": 0.54, "chosen_reward": 0.02923889271914959, "rejected_reward": -0.10640259087085724, "lr": 1.8602760313250453e-06, "tokens_per_sec": 153870.4295516376, "vram_gb": 4.5285258293151855}
|
| 157 |
+
{"timestamp": 1787715630.9617212, "elapsed_hours": 0.17857715997431015, "event": "dpo_step", "step": 1490, "loss": 0.6792966240644455, "pref_accuracy": 0.6, "chosen_reward": 0.10093536972999573, "rejected_reward": -0.077972412109375, "lr": 1.8587495908483397e-06, "tokens_per_sec": 154171.7272466924, "vram_gb": 4.074960231781006}
|
| 158 |
+
{"timestamp": 1787715634.7054539, "elapsed_hours": 0.17961708572175766, "event": "dpo_step", "step": 1500, "loss": 0.6737938326597214, "pref_accuracy": 0.64, "chosen_reward": 0.252371609210968, "rejected_reward": 0.2772003412246704, "lr": 1.8572155673277334e-06, "tokens_per_sec": 154294.44002349096, "vram_gb": 4.956308364868164}
|
| 159 |
+
{"timestamp": 1787715646.0152013, "elapsed_hours": 0.1827586821715037, "event": "dpo_step", "step": 1510, "loss": 0.6951003742218017, "pref_accuracy": 0.58, "chosen_reward": -0.004917907528579235, "rejected_reward": 0.1416061371564865, "lr": 1.8556739760944608e-06, "tokens_per_sec": 152613.96400115307, "vram_gb": 4.957822799682617}
|
| 160 |
+
{"timestamp": 1787715649.9318888, "elapsed_hours": 0.18384665091832478, "event": "dpo_step", "step": 1520, "loss": 0.695240615606308, "pref_accuracy": 0.58, "chosen_reward": -0.32677000761032104, "rejected_reward": -0.2598327696323395, "lr": 1.852568152192864e-06, "tokens_per_sec": 152702.28483536324, "vram_gb": 4.5285258293151855}
|
| 161 |
+
{"timestamp": 1787715653.3864846, "elapsed_hours": 0.1848062609963947, "event": "dpo_step", "step": 1530, "loss": 0.6985017573833465, "pref_accuracy": 0.54, "chosen_reward": 0.02592773735523224, "rejected_reward": -0.0926513671875, "lr": 1.851003950564555e-06, "tokens_per_sec": 152897.21891453804, "vram_gb": 4.962896823883057}
|
| 162 |
+
{"timestamp": 1787715657.966506, "elapsed_hours": 0.18607848909166125, "event": "dpo_step", "step": 1540, "loss": 0.7074046897888183, "pref_accuracy": 0.47, "chosen_reward": -0.06132812425494194, "rejected_reward": 0.02362806908786297, "lr": 1.8494322433033003e-06, "tokens_per_sec": 152829.20933656607, "vram_gb": 4.375131607055664}
|
| 163 |
+
{"timestamp": 1787715662.5299132, "elapsed_hours": 0.18734610206551022, "event": "dpo_step", "step": 1550, "loss": 0.6995374858379364, "pref_accuracy": 0.45, "chosen_reward": 0.10559997707605362, "rejected_reward": 0.14024430513381958, "lr": 1.8478530461169503e-06, "tokens_per_sec": 152765.94605599693, "vram_gb": 4.956491470336914}
|
| 164 |
+
{"timestamp": 1787715667.499944, "elapsed_hours": 0.18872666623857287, "event": "dpo_step", "step": 1560, "loss": 0.6822193610668182, "pref_accuracy": 0.5, "chosen_reward": 0.04633179306983948, "rejected_reward": -0.12001953274011612, "lr": 1.8446722451744861e-06, "tokens_per_sec": 152610.86998842168, "vram_gb": 4.5285258293151855}
|
| 165 |
+
{"timestamp": 1787715671.1167347, "elapsed_hours": 0.18973133034176295, "event": "dpo_step", "step": 1570, "loss": 0.6900843596458435, "pref_accuracy": 0.5, "chosen_reward": -0.14652405679225922, "rejected_reward": 0.32819825410842896, "lr": 1.8430706732077177e-06, "tokens_per_sec": 152764.38539904272, "vram_gb": 4.955942153930664}
|
| 166 |
+
{"timestamp": 1787715676.0673823, "elapsed_hours": 0.1911065102285809, "event": "dpo_step", "step": 1580, "loss": 0.6970367228984833, "pref_accuracy": 0.52, "chosen_reward": -0.11729736626148224, "rejected_reward": -0.18306885659694672, "lr": 1.8414616748942283e-06, "tokens_per_sec": 152615.6372046617, "vram_gb": 4.674154281616211}
|
| 167 |
+
{"timestamp": 1787715679.5144486, "elapsed_hours": 0.19206402864721087, "event": "dpo_step", "step": 1590, "loss": 0.6881522500514984, "pref_accuracy": 0.57, "chosen_reward": -0.01597290113568306, "rejected_reward": -0.013262942433357239, "lr": 1.8398452663145603e-06, "tokens_per_sec": 152805.2292501071, "vram_gb": 4.522758483886719}
|
| 168 |
+
{"timestamp": 1787715683.3375125, "elapsed_hours": 0.1931259908941057, "event": "dpo_step", "step": 1600, "loss": 0.7063249564170837, "pref_accuracy": 0.54, "chosen_reward": 0.024120330810546875, "rejected_reward": 0.09131889790296555, "lr": 1.836590283048991e-06, "tokens_per_sec": 152909.01795111573, "vram_gb": 3.9820127487182617}
|
| 169 |
+
{"timestamp": 1787715686.4043505, "elapsed_hours": 0.19397789027955797, "event": "sample", "step": 1600, "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 else:\n return fibonacci(n-1) + fibonacci(n-2)\n```\n\nIn this function, we use the `fibonacci()` function to calculate the nth Fibonacci number. The function returns the nth Fibonacci number.\n\nThe function returns the nth Fibonacci number, which is the number of Fibonacci numbers.<|im_end|>"}
|
| 170 |
+
{"timestamp": 1787715689.77201, "elapsed_hours": 0.19491335133711496, "event": "dpo_step", "step": 1610, "loss": 0.7063090479373932, "pref_accuracy": 0.54, "chosen_reward": 0.27839356660842896, "rejected_reward": -0.0052642822265625, "lr": 1.8349517408938238e-06, "tokens_per_sec": 152434.3818661125, "vram_gb": 4.947779655456543}
|
| 171 |
+
{"timestamp": 1787715693.2252827, "elapsed_hours": 0.19587259378698138, "event": "dpo_step", "step": 1620, "loss": 0.6969519996643067, "pref_accuracy": 0.56, "chosen_reward": -0.06976929306983948, "rejected_reward": -0.027275849133729935, "lr": 1.8333058535336204e-06, "tokens_per_sec": 152619.77267924423, "vram_gb": 4.06575345993042}
|
| 172 |
+
{"timestamp": 1787715697.2903376, "elapsed_hours": 0.19700177556938595, "event": "dpo_step", "step": 1630, "loss": 0.6876125705242156, "pref_accuracy": 0.58, "chosen_reward": -0.2906906306743622, "rejected_reward": -0.09919662773609161, "lr": 1.8316526374175968e-06, "tokens_per_sec": 152669.72087146598, "vram_gb": 4.956247329711914}
|
| 173 |
+
{"timestamp": 1787715701.3453236, "elapsed_hours": 0.1981281606356303, "event": "dpo_step", "step": 1640, "loss": 0.6916904819011688, "pref_accuracy": 0.55, "chosen_reward": 0.24091798067092896, "rejected_reward": -0.12410889565944672, "lr": 1.8283242850810132e-06, "tokens_per_sec": 152721.28377048887, "vram_gb": 4.5285258293151855}
|
| 174 |
+
{"timestamp": 1787715705.14212, "elapsed_hours": 0.19918282628059386, "event": "dpo_step", "step": 1650, "loss": 0.6880402672290802, "pref_accuracy": 0.55, "chosen_reward": -0.013940430246293545, "rejected_reward": 0.26450270414352417, "lr": 1.8266491821244464e-06, "tokens_per_sec": 152828.00500302433, "vram_gb": 4.947904586791992}
|
| 175 |
+
{"timestamp": 1787715710.359088, "elapsed_hours": 0.2006319839424557, "event": "dpo_step", "step": 1660, "loss": 0.6926880288124084, "pref_accuracy": 0.53, "chosen_reward": 0.0472412183880806, "rejected_reward": 0.073974609375, "lr": 1.8249668169397157e-06, "tokens_per_sec": 152628.868455546, "vram_gb": 4.956308364868164}
|
| 176 |
+
{"timestamp": 1787715714.5682948, "elapsed_hours": 0.20180120819144778, "event": "dpo_step", "step": 1670, "loss": 0.701442289352417, "pref_accuracy": 0.49, "chosen_reward": -0.231231689453125, "rejected_reward": 0.026715092360973358, "lr": 1.8232772063406014e-06, "tokens_per_sec": 152646.87964899983, "vram_gb": 4.40756368637085}
|
| 177 |
+
{"timestamp": 1787715718.9747827, "elapsed_hours": 0.20302523261970945, "event": "dpo_step", "step": 1680, "loss": 0.7092375516891479, "pref_accuracy": 0.47, "chosen_reward": -0.03097686544060707, "rejected_reward": -0.041030120104551315, "lr": 1.8198763165162323e-06, "tokens_per_sec": 152622.93087317268, "vram_gb": 4.212912559509277}
|
| 178 |
+
{"timestamp": 1787715723.236358, "elapsed_hours": 0.20420900351471372, "event": "dpo_step", "step": 1690, "loss": 0.7173076689243316, "pref_accuracy": 0.47, "chosen_reward": 0.06316223740577698, "rejected_reward": 0.0020843520760536194, "lr": 1.818165071279921e-06, "tokens_per_sec": 152629.74688938772, "vram_gb": 4.473140716552734}
|
| 179 |
+
{"timestamp": 1787715728.4118557, "elapsed_hours": 0.20564664171801672, "event": "dpo_step", "step": 1700, "loss": 0.7133392572402955, "pref_accuracy": 0.5, "chosen_reward": -0.05133362114429474, "rejected_reward": 0.1166786178946495, "lr": 1.8164466486067728e-06, "tokens_per_sec": 152445.60150150326, "vram_gb": 4.956125259399414}
|
| 180 |
+
{"timestamp": 1787715732.0255446, "elapsed_hours": 0.2066504442691803, "event": "dpo_step", "step": 1710, "loss": 0.7248371875286103, "pref_accuracy": 0.44, "chosen_reward": -0.1127525269985199, "rejected_reward": -0.05162787437438965, "lr": 1.8147210656709317e-06, "tokens_per_sec": 152587.8439699154, "vram_gb": 3.6783666610717773}
|
| 181 |
+
{"timestamp": 1787715735.8825765, "elapsed_hours": 0.20772184193134308, "event": "dpo_step", "step": 1720, "loss": 0.7159318244457245, "pref_accuracy": 0.45, "chosen_reward": 0.00347900390625, "rejected_reward": -0.10484619438648224, "lr": 1.8112484880653762e-06, "tokens_per_sec": 152678.3563590787, "vram_gb": 4.5285258293151855}
|
| 182 |
+
{"timestamp": 1787715740.1099365, "elapsed_hours": 0.20889610860082838, "event": "dpo_step", "step": 1730, "loss": 0.6981403386592865, "pref_accuracy": 0.48, "chosen_reward": 0.08278197795152664, "rejected_reward": 0.025609588250517845, "lr": 1.8095015281010619e-06, "tokens_per_sec": 152691.7337082502, "vram_gb": 3.8600378036499023}
|
| 183 |
+
{"timestamp": 1787715743.9060452, "elapsed_hours": 0.20995058324601915, "event": "dpo_step", "step": 1740, "loss": 0.6964376211166382, "pref_accuracy": 0.47, "chosen_reward": -0.16195908188819885, "rejected_reward": 0.1257675141096115, "lr": 1.8077474772845089e-06, "tokens_per_sec": 152793.20435334032, "vram_gb": 4.956247329711914}
|
| 184 |
+
{"timestamp": 1787715748.6206403, "elapsed_hours": 0.21126019299030305, "event": "dpo_step", "step": 1750, "loss": 0.6930470025539398, "pref_accuracy": 0.48, "chosen_reward": 0.13381347060203552, "rejected_reward": -0.08526305854320526, "lr": 1.8059863531459335e-06, "tokens_per_sec": 152706.65715865148, "vram_gb": 4.956125259399414}
|
| 185 |
+
{"timestamp": 1787715752.8851867, "elapsed_hours": 0.2124447892109553, "event": "dpo_step", "step": 1760, "loss": 0.6811084997653961, "pref_accuracy": 0.53, "chosen_reward": -0.43498536944389343, "rejected_reward": -0.4056152403354645, "lr": 1.8024429553768614e-06, "tokens_per_sec": 152712.13409105057, "vram_gb": 4.5285258293151855}
|
| 186 |
+
{"timestamp": 1787715756.9493806, "elapsed_hours": 0.21357373197873433, "event": "dpo_step", "step": 1770, "loss": 0.6816229689121246, "pref_accuracy": 0.55, "chosen_reward": 0.03847656399011612, "rejected_reward": 0.0399627685546875, "lr": 1.8006607171595515e-06, "tokens_per_sec": 152757.84162540594, "vram_gb": 4.15024995803833}
|
| 187 |
+
{"timestamp": 1787715760.0595613, "elapsed_hours": 0.21443767110506692, "event": "dpo_step", "step": 1780, "loss": 0.6941566026210785, "pref_accuracy": 0.53, "chosen_reward": 0.06125488504767418, "rejected_reward": 0.01119384728372097, "lr": 1.798871476446238e-06, "tokens_per_sec": 152994.25613406647, "vram_gb": 3.9147162437438965}
|
| 188 |
+
{"timestamp": 1787715764.7681267, "elapsed_hours": 0.21574560595883263, "event": "dpo_step", "step": 1790, "loss": 0.6800134134292602, "pref_accuracy": 0.57, "chosen_reward": -0.03200683742761612, "rejected_reward": -0.04661865159869194, "lr": 1.79707525111883e-06, "tokens_per_sec": 152909.49419553057, "vram_gb": 4.9400482177734375}
|
| 189 |
+
{"timestamp": 1787715768.8792255, "elapsed_hours": 0.21688757790459526, "event": "dpo_step", "step": 1800, "loss": 0.6733341264724731, "pref_accuracy": 0.58, "chosen_reward": -0.0546875037252903, "rejected_reward": -0.0424499548971653, "lr": 1.7934619184982102e-06, "tokens_per_sec": 152944.15891513202, "vram_gb": 4.5285258293151855}
|
| 190 |
+
{"timestamp": 1787715771.9992409, "elapsed_hours": 0.21775424877802532, "event": "sample", "step": 1800, "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 1\n elif n == 1:\n return n * 2\n else:\n return fibonacci(n-1) + fibonacci(n-2)\n```\n\nHere's an explanation of how it works:\n\n1. We start by initializing an empty list called `nums` to store the Fibonacci numbers.\n2. We iterate over each number in"}
|
| 191 |
+
{"timestamp": 1787715775.480702, "elapsed_hours": 0.2187213212913937, "event": "dpo_step", "step": 1810, "loss": 0.682735743522644, "pref_accuracy": 0.6, "chosen_reward": 0.07159423828125, "rejected_reward": 0.09308166801929474, "lr": 1.7916448473171235e-06, "tokens_per_sec": 152488.69572312656, "vram_gb": 4.944598197937012}
|
| 192 |
+
{"timestamp": 1787715780.2155528, "elapsed_hours": 0.22003655764791702, "event": "dpo_step", "step": 1820, "loss": 0.6893683457374573, "pref_accuracy": 0.57, "chosen_reward": 0.06141052022576332, "rejected_reward": 0.17779541015625, "lr": 1.7898208637458303e-06, "tokens_per_sec": 152403.5350867347, "vram_gb": 4.956125259399414}
|
| 193 |
+
{"timestamp": 1787715785.1398568, "elapsed_hours": 0.2214044198062685, "event": "dpo_step", "step": 1830, "loss": 0.6986681127548218, "pref_accuracy": 0.55, "chosen_reward": 0.1568603664636612, "rejected_reward": 0.466552734375, "lr": 1.7879899860134648e-06, "tokens_per_sec": 152282.75034407424, "vram_gb": 4.927882671356201}
|
| 194 |
+
{"timestamp": 1787715788.940331, "elapsed_hours": 0.2224601070748435, "event": "dpo_step", "step": 1840, "loss": 0.7316406178474426, "pref_accuracy": 0.46, "chosen_reward": -0.3419647216796875, "rejected_reward": 0.3232482969760895, "lr": 1.7843076213263711e-06, "tokens_per_sec": 152379.56139405203, "vram_gb": 4.5285258293151855}
|
| 195 |
+
{"timestamp": 1787715793.5863898, "elapsed_hours": 0.2237506789631314, "event": "dpo_step", "step": 1850, "loss": 0.746660737991333, "pref_accuracy": 0.41, "chosen_reward": -0.209014892578125, "rejected_reward": 0.21009674668312073, "lr": 1.7824561711736837e-06, "tokens_per_sec": 152313.48111356096, "vram_gb": 4.956308364868164}
|
| 196 |
+
{"timestamp": 1787715797.8936486, "elapsed_hours": 0.22494713975323571, "event": "dpo_step", "step": 1860, "loss": 0.7445927500724793, "pref_accuracy": 0.41, "chosen_reward": 0.3310791254043579, "rejected_reward": 0.07811279594898224, "lr": 1.7805979004636363e-06, "tokens_per_sec": 152312.61042270868, "vram_gb": 4.9539361000061035}
|
| 197 |
+
{"timestamp": 1787715802.1359026, "elapsed_hours": 0.22612554364734225, "event": "dpo_step", "step": 1870, "loss": 0.7553593325614929, "pref_accuracy": 0.42, "chosen_reward": -0.3841797113418579, "rejected_reward": 0.09450682997703552, "lr": 1.778732827768033e-06, "tokens_per_sec": 152324.06181031273, "vram_gb": 4.948909759521484}
|
| 198 |
+
{"timestamp": 1787715807.1781354, "elapsed_hours": 0.22752616392241584, "event": "dpo_step", "step": 1880, "loss": 0.7498723077774048, "pref_accuracy": 0.42, "chosen_reward": 0.19356077909469604, "rejected_reward": 0.10444336384534836, "lr": 1.7749823510470848e-06, "tokens_per_sec": 152184.8628551725, "vram_gb": 4.5285258293151855}
|
| 199 |
+
{"timestamp": 1787715810.911457, "elapsed_hours": 0.22856319771872627, "event": "dpo_step", "step": 1890, "loss": 0.7324874126911163, "pref_accuracy": 0.46, "chosen_reward": 0.03707275539636612, "rejected_reward": -0.03163452446460724, "lr": 1.7730969845045014e-06, "tokens_per_sec": 152292.08331231796, "vram_gb": 4.940404891967773}
|
| 200 |
+
{"timestamp": 1787715815.7939696, "elapsed_hours": 0.2299194512102339, "event": "dpo_step", "step": 1900, "loss": 0.7245056092739105, "pref_accuracy": 0.51, "chosen_reward": -0.03406982496380806, "rejected_reward": -0.09345092624425888, "lr": 1.771204890941509e-06, "tokens_per_sec": 152184.27280860848, "vram_gb": 4.956575393676758}
|
| 201 |
+
{"timestamp": 1787715819.9681237, "elapsed_hours": 0.23107893844445546, "event": "dpo_step", "step": 1910, "loss": 0.7286586892604828, "pref_accuracy": 0.48, "chosen_reward": 0.04536818712949753, "rejected_reward": -0.008767700754106045, "lr": 1.7693060892679416e-06, "tokens_per_sec": 152208.77968858057, "vram_gb": 4.956247329711914}
|
| 202 |
+
{"timestamp": 1787715824.2397058, "elapsed_hours": 0.23226548903518252, "event": "dpo_step", "step": 1920, "loss": 0.7124713599681854, "pref_accuracy": 0.51, "chosen_reward": 0.13860169053077698, "rejected_reward": 0.07524414360523224, "lr": 1.7654884375634355e-06, "tokens_per_sec": 152215.014268468, "vram_gb": 4.5285258293151855}
|
| 203 |
+
{"timestamp": 1787715827.3111277, "elapsed_hours": 0.23311866177452936, "event": "dpo_step", "step": 1930, "loss": 0.6990719902515411, "pref_accuracy": 0.55, "chosen_reward": -0.056017301976680756, "rejected_reward": -0.1498008817434311, "lr": 1.7635696256866136e-06, "tokens_per_sec": 152441.40924459492, "vram_gb": 4.070664882659912}
|
| 204 |
+
{"timestamp": 1787715829.9308074, "elapsed_hours": 0.23384635051091512, "event": "dpo_step", "step": 1940, "loss": 0.691464250087738, "pref_accuracy": 0.58, "chosen_reward": -0.02382659912109375, "rejected_reward": -0.003137398511171341, "lr": 1.7616441820070685e-06, "tokens_per_sec": 152748.99609849777, "vram_gb": 3.8588123321533203}
|
| 205 |
+
{"timestamp": 1787715834.6571012, "elapsed_hours": 0.23515920996665954, "event": "dpo_step", "step": 1950, "loss": 0.6802725398540497, "pref_accuracy": 0.62, "chosen_reward": -0.05427246168255806, "rejected_reward": -0.17780762910842896, "lr": 1.7597121257679395e-06, "tokens_per_sec": 152669.45566505814, "vram_gb": 4.956369400024414}
|
| 206 |
+
{"timestamp": 1787715838.6885657, "elapsed_hours": 0.236279061237971, "event": "dpo_step", "step": 1960, "loss": 0.6742738449573517, "pref_accuracy": 0.6, "chosen_reward": -0.44175416231155396, "rejected_reward": -0.08843383938074112, "lr": 1.755828252913728e-06, "tokens_per_sec": 152716.86467068034, "vram_gb": 4.5285258293151855}
|
| 207 |
+
{"timestamp": 1787715842.9179592, "elapsed_hours": 0.23745389262835184, "event": "dpo_step", "step": 1970, "loss": 0.6874585366249084, "pref_accuracy": 0.56, "chosen_reward": -0.14295348525047302, "rejected_reward": 0.032831571996212006, "lr": 1.7538764751145855e-06, "tokens_per_sec": 152728.05191711846, "vram_gb": 3.8000621795654297}
|
| 208 |
+
{"timestamp": 1787715847.1795971, "elapsed_hours": 0.23863768113984002, "event": "dpo_step", "step": 1980, "loss": 0.6929110419750214, "pref_accuracy": 0.56, "chosen_reward": -0.229705810546875, "rejected_reward": 0.2247314453125, "lr": 1.7519181623873502e-06, "tokens_per_sec": 152733.3452165379, "vram_gb": 4.956308364868164}
|
| 209 |
+
{"timestamp": 1787715851.3992965, "elapsed_hours": 0.23980981979105206, "event": "dpo_step", "step": 1990, "loss": 0.7109741926193237, "pref_accuracy": 0.51, "chosen_reward": -0.0060882568359375, "rejected_reward": -0.04886970669031143, "lr": 1.7499533343036597e-06, "tokens_per_sec": 152746.07590473193, "vram_gb": 4.476655006408691}
|
| 210 |
+
{"timestamp": 1787715856.098474, "elapsed_hours": 0.24111514680915408, "event": "dpo_step", "step": 2000, "loss": 0.7358329057693481, "pref_accuracy": 0.41, "chosen_reward": 0.0006725313141942024, "rejected_reward": 0.09378433227539062, "lr": 1.746004210678839e-06, "tokens_per_sec": 152673.36504976315, "vram_gb": 3.3267931938171387}
|
| 211 |
+
{"timestamp": 1787715859.2734165, "elapsed_hours": 0.24199707527955372, "event": "sample", "step": 2000, "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 else:\n return fibonacci(n-1) + fibonacci(n-2)\n```\n\nIn this function, the function takes an integer `n` as input. It initializes an empty list called `numbers` to store the Fibonacci numbers. Then, it iterates through each number `n` in the input `numbers` using a `for` loop. For each number, it checks if"}
|
| 212 |
+
{"timestamp": 1787715870.08541, "elapsed_hours": 0.24500040690104166, "event": "dpo_step", "step": 2010, "loss": 0.7359667205810547, "pref_accuracy": 0.43, "chosen_reward": -0.0014068633317947388, "rejected_reward": -0.28706055879592896, "lr": 1.7440199546057735e-06, "tokens_per_sec": 150977.01171137553, "vram_gb": 4.956308364868164}
|
| 213 |
+
{"timestamp": 1787715874.564904, "elapsed_hours": 0.24624471075005003, "event": "dpo_step", "step": 2020, "loss": 0.7260273778438568, "pref_accuracy": 0.42, "chosen_reward": 0.05783691257238388, "rejected_reward": 0.02709808200597763, "lr": 1.7420292621119871e-06, "tokens_per_sec": 150953.12962440812, "vram_gb": 4.952724933624268}
|
| 214 |
+
{"timestamp": 1787715878.7558017, "elapsed_hours": 0.24740884886847603, "event": "dpo_step", "step": 2030, "loss": 0.7382133841514588, "pref_accuracy": 0.38, "chosen_reward": 0.01687927171587944, "rejected_reward": 0.24624939262866974, "lr": 1.7400321530927244e-06, "tokens_per_sec": 150978.9221542875, "vram_gb": 4.955705642700195}
|
| 215 |
+
{"timestamp": 1787715882.911306, "elapsed_hours": 0.248563155664338, "event": "dpo_step", "step": 2040, "loss": 0.7236392033100129, "pref_accuracy": 0.41, "chosen_reward": 0.1148681640625, "rejected_reward": 0.088897705078125, "lr": 1.736018765379185e-06, "tokens_per_sec": 151010.51426196986, "vram_gb": 4.5285258293151855}
|
| 216 |
+
{"timestamp": 1787715887.0070243, "elapsed_hours": 0.24970085534784528, "event": "dpo_step", "step": 2050, "loss": 0.7159150385856629, "pref_accuracy": 0.41, "chosen_reward": -0.061383821070194244, "rejected_reward": 0.12672042846679688, "lr": 1.7340025267952374e-06, "tokens_per_sec": 151051.96306688333, "vram_gb": 3.948741912841797}
|
| 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}
|