Upload dpo_metrics.jsonl with huggingface_hub
Browse files- dpo_metrics.jsonl +67 -0
dpo_metrics.jsonl
CHANGED
|
@@ -732,3 +732,70 @@
|
|
| 732 |
{"timestamp": 1787719923.5364342, "elapsed_hours": 0.8351443061563704, "event": "dpo_step", "step": 6970, "loss": 0.8455632567405701, "pref_accuracy": 0.47, "chosen_reward": -0.12764282524585724, "rejected_reward": 0.11565857380628586, "lr": 3.9720061276619634e-07, "tokens_per_sec": 152385.48605476995, "vram_gb": 40.319719314575195}
|
| 733 |
{"timestamp": 1787719927.6259918, "elapsed_hours": 0.8362802944580714, "event": "dpo_step", "step": 6980, "loss": 0.8141857600212097, "pref_accuracy": 0.51, "chosen_reward": 0.17601929605007172, "rejected_reward": 0.8984527587890625, "lr": 3.94199743793352e-07, "tokens_per_sec": 152396.2055736182, "vram_gb": 41.17239284515381}
|
| 734 |
{"timestamp": 1787719932.0816817, "elapsed_hours": 0.8375179860989252, "event": "dpo_step", "step": 6990, "loss": 0.8309123855829239, "pref_accuracy": 0.49, "chosen_reward": 0.1314697563648224, "rejected_reward": -0.17182616889476776, "lr": 3.9122091988040666e-07, "tokens_per_sec": 152388.33102186298, "vram_gb": 41.15994071960449}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 732 |
{"timestamp": 1787719923.5364342, "elapsed_hours": 0.8351443061563704, "event": "dpo_step", "step": 6970, "loss": 0.8455632567405701, "pref_accuracy": 0.47, "chosen_reward": -0.12764282524585724, "rejected_reward": 0.11565857380628586, "lr": 3.9720061276619634e-07, "tokens_per_sec": 152385.48605476995, "vram_gb": 40.319719314575195}
|
| 733 |
{"timestamp": 1787719927.6259918, "elapsed_hours": 0.8362802944580714, "event": "dpo_step", "step": 6980, "loss": 0.8141857600212097, "pref_accuracy": 0.51, "chosen_reward": 0.17601929605007172, "rejected_reward": 0.8984527587890625, "lr": 3.94199743793352e-07, "tokens_per_sec": 152396.2055736182, "vram_gb": 41.17239284515381}
|
| 734 |
{"timestamp": 1787719932.0816817, "elapsed_hours": 0.8375179860989252, "event": "dpo_step", "step": 6990, "loss": 0.8309123855829239, "pref_accuracy": 0.49, "chosen_reward": 0.1314697563648224, "rejected_reward": -0.17182616889476776, "lr": 3.9122091988040666e-07, "tokens_per_sec": 152388.33102186298, "vram_gb": 41.15994071960449}
|
| 735 |
+
{"timestamp": 1787719937.5524113, "elapsed_hours": 0.8390376331408819, "event": "dpo_step", "step": 7000, "loss": 0.8149189084768296, "pref_accuracy": 0.52, "chosen_reward": 0.2107338011264801, "rejected_reward": -0.53521728515625, "lr": 3.853295260966362e-07, "tokens_per_sec": 152329.12129180745, "vram_gb": 40.74140214920044}
|
| 736 |
+
{"timestamp": 1787719940.6494758, "elapsed_hours": 0.8398979289664162, "event": "sample", "step": 7000, "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 else:\n return fibonacci(n-1) + fibonacci(n-2)\n```\n\nThis function first checks if `n` is equal to 0. If it is, it returns 0 as the first Fibonacci number. If it is not, it returns 1 as the second Fibonacci number.\n\n"}
|
| 737 |
+
{"timestamp": 1787719951.158002, "elapsed_hours": 0.8428169639243019, "event": "dpo_step", "step": 7010, "loss": 0.7779387736320496, "pref_accuracy": 0.5, "chosen_reward": -0.42253419756889343, "rejected_reward": -0.9299377799034119, "lr": 3.82417015105181e-07, "tokens_per_sec": 151860.6654605726, "vram_gb": 40.78501844406128}
|
| 738 |
+
{"timestamp": 1787719955.11474, "elapsed_hours": 0.8439160577456156, "event": "dpo_step", "step": 7020, "loss": 0.7371939986944198, "pref_accuracy": 0.49, "chosen_reward": -0.508374035358429, "rejected_reward": 0.06034699082374573, "lr": 3.7952666693177567e-07, "tokens_per_sec": 151878.6524454647, "vram_gb": 41.158527851104736}
|
| 739 |
+
{"timestamp": 1787719959.1226175, "elapsed_hours": 0.845029357208146, "event": "dpo_step", "step": 7030, "loss": 0.7364758121967315, "pref_accuracy": 0.48, "chosen_reward": 0.15100708603858948, "rejected_reward": 0.2076416313648224, "lr": 3.76658510462943e-07, "tokens_per_sec": 151894.0323900516, "vram_gb": 41.160499572753906}
|
| 740 |
+
{"timestamp": 1787719962.0310047, "elapsed_hours": 0.8458372424708472, "event": "dpo_step", "step": 7040, "loss": 0.6930319398641587, "pref_accuracy": 0.57, "chosen_reward": 0.12420044839382172, "rejected_reward": -0.0011043548583984375, "lr": 3.709888870758624e-07, "tokens_per_sec": 151964.38400401745, "vram_gb": 39.59065008163452}
|
| 741 |
+
{"timestamp": 1787719965.048229, "elapsed_hours": 0.846675360335244, "event": "dpo_step", "step": 7050, "loss": 0.7293305373191834, "pref_accuracy": 0.52, "chosen_reward": 0.24407958984375, "rejected_reward": -0.43216782808303833, "lr": 3.6818747682058267e-07, "tokens_per_sec": 152029.1577205714, "vram_gb": 40.144402503967285}
|
| 742 |
+
{"timestamp": 1787719968.8400621, "elapsed_hours": 0.847728647324774, "event": "dpo_step", "step": 7060, "loss": 0.753392915725708, "pref_accuracy": 0.55, "chosen_reward": -0.12756958603858948, "rejected_reward": 0.17534179985523224, "lr": 3.6540837159524225e-07, "tokens_per_sec": 152055.08604156764, "vram_gb": 41.15110635757446}
|
| 743 |
+
{"timestamp": 1787719972.723263, "elapsed_hours": 0.8488073142369589, "event": "dpo_step", "step": 7070, "loss": 0.7323194688558579, "pref_accuracy": 0.61, "chosen_reward": -0.13350892066955566, "rejected_reward": -0.052808769047260284, "lr": 3.6265159917458854e-07, "tokens_per_sec": 152076.38535402998, "vram_gb": 39.946998596191406}
|
| 744 |
+
{"timestamp": 1787719976.7346566, "elapsed_hours": 0.8499215901560253, "event": "dpo_step", "step": 7080, "loss": 0.7068795174360275, "pref_accuracy": 0.61, "chosen_reward": 0.045025646686553955, "rejected_reward": -0.3258728086948395, "lr": 3.5720516273005895e-07, "tokens_per_sec": 152091.2404914486, "vram_gb": 40.74140214920044}
|
| 745 |
+
{"timestamp": 1787719981.011021, "elapsed_hours": 0.8511094691356024, "event": "dpo_step", "step": 7090, "loss": 0.7404549241065979, "pref_accuracy": 0.53, "chosen_reward": 0.6456894278526306, "rejected_reward": -0.047246936708688736, "lr": 3.5451555313859157e-07, "tokens_per_sec": 152092.86497920737, "vram_gb": 41.17890930175781}
|
| 746 |
+
{"timestamp": 1787719984.1495445, "elapsed_hours": 0.8519812813070086, "event": "dpo_step", "step": 7100, "loss": 0.7483466356992722, "pref_accuracy": 0.55, "chosen_reward": -0.18546754121780396, "rejected_reward": 0.0028198063373565674, "lr": 3.5184838521608026e-07, "tokens_per_sec": 152151.07330154095, "vram_gb": 40.29124069213867}
|
| 747 |
+
{"timestamp": 1787719988.262476, "elapsed_hours": 0.853123762342665, "event": "dpo_step", "step": 7110, "loss": 0.7391969674825668, "pref_accuracy": 0.56, "chosen_reward": -0.20810624957084656, "rejected_reward": -0.2861572504043579, "lr": 3.492036856185542e-07, "tokens_per_sec": 152160.7334205738, "vram_gb": 40.578256607055664}
|
| 748 |
+
{"timestamp": 1787719991.1163628, "elapsed_hours": 0.8539165086216397, "event": "dpo_step", "step": 7120, "loss": 0.7507239216566086, "pref_accuracy": 0.52, "chosen_reward": -0.17859192192554474, "rejected_reward": 0.05115967243909836, "lr": 3.43981796899555e-07, "tokens_per_sec": 152232.86915118236, "vram_gb": 40.74140214920044}
|
| 749 |
+
{"timestamp": 1787719995.7322, "elapsed_hours": 0.8551986855930752, "event": "dpo_step", "step": 7130, "loss": 0.7835756903886795, "pref_accuracy": 0.53, "chosen_reward": 0.6688293814659119, "rejected_reward": -0.23920214176177979, "lr": 3.414046599663295e-07, "tokens_per_sec": 152217.45309738378, "vram_gb": 41.155048847198486}
|
| 750 |
+
{"timestamp": 1787719999.7524998, "elapsed_hours": 0.8563154355684917, "event": "dpo_step", "step": 7140, "loss": 0.7855291980504989, "pref_accuracy": 0.56, "chosen_reward": 0.8767788410186768, "rejected_reward": 0.12515830993652344, "lr": 3.3885009573406275e-07, "tokens_per_sec": 152231.5728677208, "vram_gb": 41.17798328399658}
|
| 751 |
+
{"timestamp": 1787720004.4744256, "elapsed_hours": 0.8576270815398958, "event": "dpo_step", "step": 7150, "loss": 0.7458832317590713, "pref_accuracy": 0.56, "chosen_reward": 0.30970460176467896, "rejected_reward": 0.6406128406524658, "lr": 3.363181297334082e-07, "tokens_per_sec": 152210.95559955234, "vram_gb": 40.90116310119629}
|
| 752 |
+
{"timestamp": 1787720008.0248172, "elapsed_hours": 0.8586133015155792, "event": "dpo_step", "step": 7160, "loss": 0.7302902001142502, "pref_accuracy": 0.56, "chosen_reward": -0.02450866997241974, "rejected_reward": -0.035980224609375, "lr": 3.313220934200468e-07, "tokens_per_sec": 152248.25459598462, "vram_gb": 40.24661636352539}
|
| 753 |
+
{"timestamp": 1787720011.3425953, "elapsed_hours": 0.8595349065462748, "event": "dpo_step", "step": 7170, "loss": 0.7284899455308914, "pref_accuracy": 0.59, "chosen_reward": 0.26129150390625, "rejected_reward": -0.4319160580635071, "lr": 3.288580730383887e-07, "tokens_per_sec": 152296.94552522872, "vram_gb": 40.24065113067627}
|
| 754 |
+
{"timestamp": 1787720016.4290936, "elapsed_hours": 0.8609478227297465, "event": "dpo_step", "step": 7180, "loss": 0.7600412827730179, "pref_accuracy": 0.56, "chosen_reward": -0.3603927791118622, "rejected_reward": 0.5042816400527954, "lr": 3.264167507499407e-07, "tokens_per_sec": 152258.3444330709, "vram_gb": 41.15547561645508}
|
| 755 |
+
{"timestamp": 1787720019.837836, "elapsed_hours": 0.8618946956263648, "event": "dpo_step", "step": 7190, "loss": 0.7482868438959122, "pref_accuracy": 0.56, "chosen_reward": 0.13337861001491547, "rejected_reward": -0.13567659258842468, "lr": 3.2399815095360155e-07, "tokens_per_sec": 152302.41466026122, "vram_gb": 41.16566467285156}
|
| 756 |
+
{"timestamp": 1787720023.419526, "elapsed_hours": 0.8628896094693078, "event": "dpo_step", "step": 7200, "loss": 0.7750080680847168, "pref_accuracy": 0.53, "chosen_reward": -0.0783691555261612, "rejected_reward": -0.6911499500274658, "lr": 3.1922921529714174e-07, "tokens_per_sec": 152337.88176263607, "vram_gb": 40.74140214920044}
|
| 757 |
+
{"timestamp": 1787720026.7353375, "elapsed_hours": 0.8638106682565477, "event": "sample", "step": 7200, "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. If `n` is less than 1, it returns `n`. Otherwise, it returns `n-1`.\n\nThe function then uses the `fibonacci` function to calculate the Fibonacci numbers. The"}
|
| 758 |
+
{"timestamp": 1787720030.2229776, "elapsed_hours": 0.8647794571850035, "event": "dpo_step", "step": 7210, "loss": 0.797348471879959, "pref_accuracy": 0.51, "chosen_reward": -0.38005372881889343, "rejected_reward": 0.5112510919570923, "lr": 3.1687892709839636e-07, "tokens_per_sec": 152215.1304455108, "vram_gb": 41.16067314147949}
|
| 759 |
+
{"timestamp": 1787720035.0379326, "elapsed_hours": 0.8661169446839226, "event": "dpo_step", "step": 7220, "loss": 0.7979761266708374, "pref_accuracy": 0.48, "chosen_reward": 0.2629966735839844, "rejected_reward": 0.3931594789028168, "lr": 3.1455145671403163e-07, "tokens_per_sec": 152190.18476699106, "vram_gb": 41.155468463897705}
|
| 760 |
+
{"timestamp": 1787720038.2928824, "elapsed_hours": 0.8670210974084006, "event": "dpo_step", "step": 7230, "loss": 0.7417602622509003, "pref_accuracy": 0.52, "chosen_reward": -0.2626487910747528, "rejected_reward": 0.18272553384304047, "lr": 3.1224682740509614e-07, "tokens_per_sec": 152241.5893475009, "vram_gb": 40.19777774810791}
|
| 761 |
+
{"timestamp": 1787720042.0189962, "elapsed_hours": 0.8680561290184656, "event": "dpo_step", "step": 7240, "loss": 0.7221362739801407, "pref_accuracy": 0.53, "chosen_reward": 0.523571789264679, "rejected_reward": 0.000274658203125, "lr": 3.077061839160853e-07, "tokens_per_sec": 152269.85806826578, "vram_gb": 40.74140214920044}
|
| 762 |
+
{"timestamp": 1787720046.560731, "elapsed_hours": 0.8693177219231923, "event": "dpo_step", "step": 7250, "loss": 0.7053551954030991, "pref_accuracy": 0.55, "chosen_reward": -0.06663207709789276, "rejected_reward": 0.007147982716560364, "lr": 3.0547021511580336e-07, "tokens_per_sec": 152258.2560197546, "vram_gb": 41.15491771697998}
|
| 763 |
+
{"timestamp": 1787720050.2490082, "elapsed_hours": 0.8703422434462441, "event": "dpo_step", "step": 7260, "loss": 0.7169129908084869, "pref_accuracy": 0.54, "chosen_reward": 0.7587753534317017, "rejected_reward": 0.4723663628101349, "lr": 3.032571781500809e-07, "tokens_per_sec": 152288.2745741595, "vram_gb": 40.082534313201904}
|
| 764 |
+
{"timestamp": 1787720054.113701, "elapsed_hours": 0.8714157691929075, "event": "dpo_step", "step": 7270, "loss": 0.7209849226474762, "pref_accuracy": 0.54, "chosen_reward": 0.27235108613967896, "rejected_reward": -0.11427612602710724, "lr": 3.0106709513630267e-07, "tokens_per_sec": 152309.63466788345, "vram_gb": 41.16221857070923}
|
| 765 |
+
{"timestamp": 1787720058.25113, "elapsed_hours": 0.8725650550921759, "event": "dpo_step", "step": 7280, "loss": 0.747504529953003, "pref_accuracy": 0.53, "chosen_reward": 0.4946594536304474, "rejected_reward": -0.06029663607478142, "lr": 2.967558782868748e-07, "tokens_per_sec": 152317.6753543224, "vram_gb": 40.74140214920044}
|
| 766 |
+
{"timestamp": 1787720062.3759015, "elapsed_hours": 0.8737108249134488, "event": "dpo_step", "step": 7290, "loss": 0.8032336002588272, "pref_accuracy": 0.48, "chosen_reward": 0.09378661215305328, "rejected_reward": 1.2134766578674316, "lr": 2.9463478753809704e-07, "tokens_per_sec": 152326.3117117524, "vram_gb": 41.16136646270752}
|
| 767 |
+
{"timestamp": 1787720066.669388, "elapsed_hours": 0.8749034600125418, "event": "dpo_step", "step": 7300, "loss": 0.7827127200365066, "pref_accuracy": 0.51, "chosen_reward": 0.08178406208753586, "rejected_reward": -0.43503114581108093, "lr": 2.925367369145777e-07, "tokens_per_sec": 152326.7416339981, "vram_gb": 41.14677619934082}
|
| 768 |
+
{"timestamp": 1787720070.26365, "elapsed_hours": 0.8759018662240771, "event": "dpo_step", "step": 7310, "loss": 0.7643106973171234, "pref_accuracy": 0.53, "chosen_reward": -0.12731324136257172, "rejected_reward": -0.03358153998851776, "lr": 2.9046174738451204e-07, "tokens_per_sec": 152361.0443129475, "vram_gb": 41.09214782714844}
|
| 769 |
+
{"timestamp": 1787720074.1447833, "elapsed_hours": 0.8769799586799409, "event": "dpo_step", "step": 7320, "loss": 0.8097970026731491, "pref_accuracy": 0.51, "chosen_reward": -0.0724056288599968, "rejected_reward": 0.18945065140724182, "lr": 2.863810343249403e-07, "tokens_per_sec": 152381.38482317232, "vram_gb": 39.73233652114868}
|
| 770 |
+
{"timestamp": 1787720077.1347587, "elapsed_hours": 0.8778105074167252, "event": "dpo_step", "step": 7330, "loss": 0.7967576795816421, "pref_accuracy": 0.48, "chosen_reward": 0.03160400688648224, "rejected_reward": -0.028491228818893433, "lr": 2.8437535157862116e-07, "tokens_per_sec": 152444.76911276372, "vram_gb": 41.15688467025757}
|
| 771 |
+
{"timestamp": 1787720080.5503473, "elapsed_hours": 0.8787592820988761, "event": "dpo_step", "step": 7340, "loss": 0.7663275307416916, "pref_accuracy": 0.49, "chosen_reward": -0.579418957233429, "rejected_reward": -0.017248541116714478, "lr": 2.8239281149172175e-07, "tokens_per_sec": 152487.4592887702, "vram_gb": 41.15413951873779}
|
| 772 |
+
{"timestamp": 1787720084.9576936, "elapsed_hours": 0.8799835450119442, "event": "dpo_step", "step": 7350, "loss": 0.7737333452701569, "pref_accuracy": 0.49, "chosen_reward": -0.01961974799633026, "rejected_reward": 0.07652205228805542, "lr": 2.804334338780106e-07, "tokens_per_sec": 152482.17049356637, "vram_gb": 41.15682029724121}
|
| 773 |
+
{"timestamp": 1787720089.4632196, "elapsed_hours": 0.8812350798977746, "event": "dpo_step", "step": 7360, "loss": 0.772362220287323, "pref_accuracy": 0.49, "chosen_reward": 0.1358184814453125, "rejected_reward": -0.06579895317554474, "lr": 2.765842441675848e-07, "tokens_per_sec": 152472.16514096264, "vram_gb": 39.91525888442993}
|
| 774 |
+
{"timestamp": 1787720092.7185867, "elapsed_hours": 0.8821393485864003, "event": "dpo_step", "step": 7370, "loss": 0.7293664890527726, "pref_accuracy": 0.5, "chosen_reward": 0.152252197265625, "rejected_reward": -0.16197510063648224, "lr": 2.746944705401827e-07, "tokens_per_sec": 152522.37892268875, "vram_gb": 40.28615713119507}
|
| 775 |
+
{"timestamp": 1787720095.5752478, "elapsed_hours": 0.8829328656196594, "event": "dpo_step", "step": 7380, "loss": 0.6987231490015984, "pref_accuracy": 0.55, "chosen_reward": 0.0761341080069542, "rejected_reward": -0.2112753987312317, "lr": 2.72827936324207e-07, "tokens_per_sec": 152591.67739236323, "vram_gb": 40.06131649017334}
|
| 776 |
+
{"timestamp": 1787720100.144845, "elapsed_hours": 0.8842021980550554, "event": "dpo_step", "step": 7390, "loss": 0.7137103292346001, "pref_accuracy": 0.56, "chosen_reward": 0.4348869323730469, "rejected_reward": -0.6651707291603088, "lr": 2.709846601740491e-07, "tokens_per_sec": 152578.4707388839, "vram_gb": 41.1680793762207}
|
| 777 |
+
{"timestamp": 1787720102.4592955, "elapsed_hours": 0.8848451010386149, "event": "dpo_step", "step": 7400, "loss": 0.7101130983233452, "pref_accuracy": 0.55, "chosen_reward": 0.2255493402481079, "rejected_reward": 0.24484863877296448, "lr": 2.673679555263401e-07, "tokens_per_sec": 152673.61340351807, "vram_gb": 40.7166633605957}
|
| 778 |
+
{"timestamp": 1787720105.6231885, "elapsed_hours": 0.8857239603334003, "event": "sample", "step": 7400, "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 2\n else:\n return fibonacci(n-1) + fibonacci(n-2)\n```\n\nHere's an explanation of the code:\n\n* The function `fibonacci` takes a single argument `n` (the input number) and returns the nth Fibonacci number.\n* The function `"}
|
| 779 |
+
{"timestamp": 1787720108.5456157, "elapsed_hours": 0.8865357455280092, "event": "dpo_step", "step": 7410, "loss": 0.7100579741597176, "pref_accuracy": 0.54, "chosen_reward": -0.2542304992675781, "rejected_reward": 0.11896362155675888, "lr": 2.6559456317461486e-07, "tokens_per_sec": 152587.56387818104, "vram_gb": 40.02397680282593}
|
| 780 |
+
{"timestamp": 1787720111.6648316, "elapsed_hours": 0.8874021944734786, "event": "dpo_step", "step": 7420, "loss": 0.6907983162999153, "pref_accuracy": 0.57, "chosen_reward": -0.358969122171402, "rejected_reward": -0.3062027096748352, "lr": 2.6384450117999607e-07, "tokens_per_sec": 152643.88008567799, "vram_gb": 40.32664155960083}
|
| 781 |
+
{"timestamp": 1787720116.4016051, "elapsed_hours": 0.8887179648213916, "event": "dpo_step", "step": 7430, "loss": 0.6966595447063446, "pref_accuracy": 0.55, "chosen_reward": -0.7590271234512329, "rejected_reward": -0.25339967012405396, "lr": 2.6211778703283485e-07, "tokens_per_sec": 152622.66661277565, "vram_gb": 41.15682029724121}
|
| 782 |
+
{"timestamp": 1787720119.670292, "elapsed_hours": 0.8896259333027734, "event": "dpo_step", "step": 7440, "loss": 0.6672569048404694, "pref_accuracy": 0.58, "chosen_reward": -0.18588009476661682, "rejected_reward": -0.09141656011343002, "lr": 2.5873447107541204e-07, "tokens_per_sec": 152671.66364812054, "vram_gb": 39.83954191207886}
|
| 783 |
+
{"timestamp": 1787720122.9711928, "elapsed_hours": 0.8905428502957026, "event": "dpo_step", "step": 7450, "loss": 0.6790256041288376, "pref_accuracy": 0.55, "chosen_reward": -0.38372498750686646, "rejected_reward": -0.1185653805732727, "lr": 2.5707790307845784e-07, "tokens_per_sec": 152719.02247705305, "vram_gb": 40.21579122543335}
|
| 784 |
+
{"timestamp": 1787720126.6575866, "elapsed_hours": 0.8915668485562006, "event": "dpo_step", "step": 7460, "loss": 0.6817837005853653, "pref_accuracy": 0.54, "chosen_reward": -0.052017975598573685, "rejected_reward": 0.3058868646621704, "lr": 2.554447505552391e-07, "tokens_per_sec": 152747.88763326715, "vram_gb": 41.16117334365845}
|
| 785 |
+
{"timestamp": 1787720130.032472, "elapsed_hours": 0.8925043166346021, "event": "dpo_step", "step": 7470, "loss": 0.6968231213092804, "pref_accuracy": 0.54, "chosen_reward": -0.5331482291221619, "rejected_reward": 0.18376466631889343, "lr": 2.538350298276985e-07, "tokens_per_sec": 152791.53564143134, "vram_gb": 41.159125328063965}
|
| 786 |
+
{"timestamp": 1787720134.1408846, "elapsed_hours": 0.8936455425288942, "event": "dpo_step", "step": 7480, "loss": 0.7381596553325653, "pref_accuracy": 0.52, "chosen_reward": -0.14458008110523224, "rejected_reward": 0.16012346744537354, "lr": 2.506859478763589e-07, "tokens_per_sec": 152800.14537460764, "vram_gb": 40.74140214920044}
|
| 787 |
+
{"timestamp": 1787720139.1211014, "elapsed_hours": 0.8950289360019896, "event": "dpo_step", "step": 7490, "loss": 0.7845041942596436, "pref_accuracy": 0.48, "chosen_reward": -0.4554497003555298, "rejected_reward": -0.0005155559629201889, "lr": 2.49146618124902e-07, "tokens_per_sec": 152767.27808159695, "vram_gb": 41.15942716598511}
|
| 788 |
+
{"timestamp": 1787720143.4049199, "elapsed_hours": 0.8962188855806986, "event": "dpo_step", "step": 7500, "loss": 0.7985517740249634, "pref_accuracy": 0.5, "chosen_reward": -0.49939730763435364, "rejected_reward": -0.11691741645336151, "lr": 2.476307831134927e-07, "tokens_per_sec": 152767.56647971142, "vram_gb": 40.457486152648926}
|
| 789 |
+
{"timestamp": 1787720185.7600806, "elapsed_hours": 0.9079842080010309, "event": "dpo_step", "step": 7510, "loss": 0.7676076197624206, "pref_accuracy": 0.53, "chosen_reward": 0.03321761637926102, "rejected_reward": 0.12623290717601776, "lr": 2.461384579915874e-07, "tokens_per_sec": 150983.66817763887, "vram_gb": 40.4295072555542}
|
| 790 |
+
{"timestamp": 1787720190.5148852, "elapsed_hours": 0.9093049870596992, "event": "dpo_step", "step": 7520, "loss": 0.7769002306461334, "pref_accuracy": 0.5, "chosen_reward": 0.02915496751666069, "rejected_reward": 0.2550613284111023, "lr": 2.432243968391564e-07, "tokens_per_sec": 150964.51128865525, "vram_gb": 39.85303497314453}
|
| 791 |
+
{"timestamp": 1787720195.3960137, "elapsed_hours": 0.9106608560350207, "event": "dpo_step", "step": 7530, "loss": 0.7847054097056388, "pref_accuracy": 0.5, "chosen_reward": 0.015179440379142761, "rejected_reward": 0.6247284412384033, "lr": 2.418026899321447e-07, "tokens_per_sec": 150939.5790046044, "vram_gb": 41.17365837097168}
|
| 792 |
+
{"timestamp": 1787720199.5376854, "elapsed_hours": 0.9118113205167983, "event": "dpo_step", "step": 7540, "loss": 0.7378646084666252, "pref_accuracy": 0.54, "chosen_reward": -0.20068359375, "rejected_reward": 0.35159456729888916, "lr": 2.4040455116137105e-07, "tokens_per_sec": 150948.8098305975, "vram_gb": 40.2755331993103}
|
| 793 |
+
{"timestamp": 1787720202.9092224, "elapsed_hours": 0.9127478584316042, "event": "dpo_step", "step": 7550, "loss": 0.7031495395302773, "pref_accuracy": 0.59, "chosen_reward": 0.13711395859718323, "rejected_reward": -0.45824891328811646, "lr": 2.390299945000205e-07, "tokens_per_sec": 150993.4958946531, "vram_gb": 41.16321802139282}
|
| 794 |
+
{"timestamp": 1787720206.2073348, "elapsed_hours": 0.9136640008290609, "event": "dpo_step", "step": 7560, "loss": 0.7391508248448372, "pref_accuracy": 0.59, "chosen_reward": 1.10467529296875, "rejected_reward": -0.10557861626148224, "lr": 2.3635168221977274e-07, "tokens_per_sec": 151041.4682898791, "vram_gb": 40.74140214920044}
|
| 795 |
+
{"timestamp": 1787720209.9484076, "elapsed_hours": 0.9147031877438228, "event": "dpo_step", "step": 7570, "loss": 0.7530837282538414, "pref_accuracy": 0.59, "chosen_reward": 0.023034662008285522, "rejected_reward": 0.23781737685203552, "lr": 2.3504795336828356e-07, "tokens_per_sec": 151068.96723340507, "vram_gb": 41.00313758850098}
|
| 796 |
+
{"timestamp": 1787720212.2365618, "elapsed_hours": 0.9153387860457103, "event": "dpo_step", "step": 7580, "loss": 0.701298548579216, "pref_accuracy": 0.62, "chosen_reward": 0.21580223739147186, "rejected_reward": -0.24277453124523163, "lr": 2.3376786016076466e-07, "tokens_per_sec": 151163.20416370017, "vram_gb": 40.3734393119812}
|
| 797 |
+
{"timestamp": 1787720216.3972352, "elapsed_hours": 0.9164945286512375, "event": "dpo_step", "step": 7590, "loss": 0.6883704003691673, "pref_accuracy": 0.61, "chosen_reward": 0.07346192002296448, "rejected_reward": -0.10067138075828552, "lr": 2.325114153906373e-07, "tokens_per_sec": 151171.23437708055, "vram_gb": 41.15795421600342}
|
| 798 |
+
{"timestamp": 1787720220.6847563, "elapsed_hours": 0.9176855068074332, "event": "dpo_step", "step": 7600, "loss": 0.7505589804053306, "pref_accuracy": 0.55, "chosen_reward": 0.029478836804628372, "rejected_reward": 0.13264617323875427, "lr": 2.3006952115438838e-07, "tokens_per_sec": 151173.42125080427, "vram_gb": 39.63038110733032}
|
| 799 |
+
{"timestamp": 1787720223.7428653, "elapsed_hours": 0.9185349815421634, "event": "sample", "step": 7600, "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# Example usage\nfibonacci(10)\n```\n\nThis function takes an integer `n` as input and returns the nth Fibonacci number.\n\nTo use this function, simply call it with an integer as an argument. For example, `fibonacci(10)` will return the 10"}
|
| 800 |
+
{"timestamp": 1787720227.6795447, "elapsed_hours": 0.9196285035212834, "event": "dpo_step", "step": 7610, "loss": 0.7444390788674354, "pref_accuracy": 0.55, "chosen_reward": -0.14771118760108948, "rejected_reward": -0.03736153244972229, "lr": 2.288840960928809e-07, "tokens_per_sec": 151051.64727778244, "vram_gb": 40.502015590667725}
|
| 801 |
+
{"timestamp": 1787720232.8000653, "elapsed_hours": 0.9210508704185486, "event": "dpo_step", "step": 7620, "loss": 0.7362357285618782, "pref_accuracy": 0.57, "chosen_reward": 0.6414734125137329, "rejected_reward": 0.1552860289812088, "lr": 2.2772236827775004e-07, "tokens_per_sec": 151015.93407523213, "vram_gb": 41.15975570678711}
|