| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9982631930527722, | |
| "eval_steps": 400, | |
| "global_step": 467, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01068804275217101, | |
| "grad_norm": 7.986297905104731, | |
| "learning_rate": 5.3191489361702123e-08, | |
| "logits/chosen": -0.48609334230422974, | |
| "logits/rejected": -0.46484899520874023, | |
| "logps/chosen": -0.7340937256813049, | |
| "logps/rejected": -0.7040716409683228, | |
| "loss": 1.7354, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.7340937256813049, | |
| "rewards/margins": -0.03002200648188591, | |
| "rewards/rejected": -0.7040716409683228, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.02137608550434202, | |
| "grad_norm": 12.558284529235404, | |
| "learning_rate": 1.0638297872340425e-07, | |
| "logits/chosen": -0.4926098883152008, | |
| "logits/rejected": -0.4637380540370941, | |
| "logps/chosen": -0.6990992426872253, | |
| "logps/rejected": -0.6799914836883545, | |
| "loss": 1.7184, | |
| "rewards/accuracies": 0.42500001192092896, | |
| "rewards/chosen": -0.6990992426872253, | |
| "rewards/margins": -0.019107775762677193, | |
| "rewards/rejected": -0.6799914836883545, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03206412825651302, | |
| "grad_norm": 8.563346586751145, | |
| "learning_rate": 1.5957446808510638e-07, | |
| "logits/chosen": -0.5024669766426086, | |
| "logits/rejected": -0.4827190339565277, | |
| "logps/chosen": -0.7477884292602539, | |
| "logps/rejected": -0.7482166886329651, | |
| "loss": 1.7413, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.7477884292602539, | |
| "rewards/margins": 0.00042818710790015757, | |
| "rewards/rejected": -0.7482166886329651, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04275217100868404, | |
| "grad_norm": 15.831378697722434, | |
| "learning_rate": 2.127659574468085e-07, | |
| "logits/chosen": -0.55638587474823, | |
| "logits/rejected": -0.5376683473587036, | |
| "logps/chosen": -0.7424782514572144, | |
| "logps/rejected": -0.767365574836731, | |
| "loss": 1.7317, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.7424782514572144, | |
| "rewards/margins": 0.024887343868613243, | |
| "rewards/rejected": -0.767365574836731, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.053440213760855046, | |
| "grad_norm": 4.334123925999533, | |
| "learning_rate": 2.659574468085106e-07, | |
| "logits/chosen": -0.5365598797798157, | |
| "logits/rejected": -0.514229953289032, | |
| "logps/chosen": -0.6908692717552185, | |
| "logps/rejected": -0.723260223865509, | |
| "loss": 1.6889, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.6908692717552185, | |
| "rewards/margins": 0.0323910117149353, | |
| "rewards/rejected": -0.723260223865509, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06412825651302605, | |
| "grad_norm": 24.739834421083724, | |
| "learning_rate": 3.1914893617021275e-07, | |
| "logits/chosen": -0.5648776292800903, | |
| "logits/rejected": -0.5341423749923706, | |
| "logps/chosen": -0.756083071231842, | |
| "logps/rejected": -0.7850486040115356, | |
| "loss": 1.7234, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.756083071231842, | |
| "rewards/margins": 0.028965547680854797, | |
| "rewards/rejected": -0.7850486040115356, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07481629926519706, | |
| "grad_norm": 10.555677691326286, | |
| "learning_rate": 3.7234042553191484e-07, | |
| "logits/chosen": -0.5409075617790222, | |
| "logits/rejected": -0.5266342759132385, | |
| "logps/chosen": -0.7565469741821289, | |
| "logps/rejected": -0.768386960029602, | |
| "loss": 1.7334, | |
| "rewards/accuracies": 0.4625000059604645, | |
| "rewards/chosen": -0.7565469741821289, | |
| "rewards/margins": 0.01184004358947277, | |
| "rewards/rejected": -0.768386960029602, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.08550434201736808, | |
| "grad_norm": 4.539232614377402, | |
| "learning_rate": 4.25531914893617e-07, | |
| "logits/chosen": -0.5185666680335999, | |
| "logits/rejected": -0.48464006185531616, | |
| "logps/chosen": -0.7097610235214233, | |
| "logps/rejected": -0.7194214463233948, | |
| "loss": 1.7191, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.7097610235214233, | |
| "rewards/margins": 0.009660452604293823, | |
| "rewards/rejected": -0.7194214463233948, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09619238476953908, | |
| "grad_norm": 7.371009874360278, | |
| "learning_rate": 4.787234042553192e-07, | |
| "logits/chosen": -0.5488617420196533, | |
| "logits/rejected": -0.5185669660568237, | |
| "logps/chosen": -0.7323862314224243, | |
| "logps/rejected": -0.7507281303405762, | |
| "loss": 1.7259, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.7323862314224243, | |
| "rewards/margins": 0.018341872841119766, | |
| "rewards/rejected": -0.7507281303405762, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10688042752171009, | |
| "grad_norm": 4.30006650630765, | |
| "learning_rate": 4.999370587356267e-07, | |
| "logits/chosen": -0.4964373707771301, | |
| "logits/rejected": -0.4881821274757385, | |
| "logps/chosen": -0.7218765020370483, | |
| "logps/rejected": -0.7749972343444824, | |
| "loss": 1.7154, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.7218765020370483, | |
| "rewards/margins": 0.0531206950545311, | |
| "rewards/rejected": -0.7749972343444824, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11756847027388109, | |
| "grad_norm": 6.20027372721574, | |
| "learning_rate": 4.995525324419337e-07, | |
| "logits/chosen": -0.5504491925239563, | |
| "logits/rejected": -0.5234394669532776, | |
| "logps/chosen": -0.7093667387962341, | |
| "logps/rejected": -0.7445443272590637, | |
| "loss": 1.7063, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.7093667387962341, | |
| "rewards/margins": 0.03517753630876541, | |
| "rewards/rejected": -0.7445443272590637, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.1282565130260521, | |
| "grad_norm": 6.6605956431585485, | |
| "learning_rate": 4.988189843662815e-07, | |
| "logits/chosen": -0.5331617593765259, | |
| "logits/rejected": -0.5232412219047546, | |
| "logps/chosen": -0.7044564485549927, | |
| "logps/rejected": -0.7688254714012146, | |
| "loss": 1.7116, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.7044564485549927, | |
| "rewards/margins": 0.06436899304389954, | |
| "rewards/rejected": -0.7688254714012146, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.13894455577822312, | |
| "grad_norm": 27.166294194610515, | |
| "learning_rate": 4.977374404419837e-07, | |
| "logits/chosen": -0.5088934302330017, | |
| "logits/rejected": -0.5036636590957642, | |
| "logps/chosen": -0.7347576022148132, | |
| "logps/rejected": -0.7716769576072693, | |
| "loss": 1.7247, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.7347576022148132, | |
| "rewards/margins": 0.03691936656832695, | |
| "rewards/rejected": -0.7716769576072693, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.14963259853039412, | |
| "grad_norm": 8.650901425425875, | |
| "learning_rate": 4.963094133060148e-07, | |
| "logits/chosen": -0.5395575761795044, | |
| "logits/rejected": -0.5354547500610352, | |
| "logps/chosen": -0.7364605069160461, | |
| "logps/rejected": -0.812324047088623, | |
| "loss": 1.7112, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.7364605069160461, | |
| "rewards/margins": 0.07586357742547989, | |
| "rewards/rejected": -0.812324047088623, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.16032064128256512, | |
| "grad_norm": 17.716918104967913, | |
| "learning_rate": 4.945369001834514e-07, | |
| "logits/chosen": -0.4694588780403137, | |
| "logits/rejected": -0.44899359345436096, | |
| "logps/chosen": -0.7442446947097778, | |
| "logps/rejected": -0.8212822675704956, | |
| "loss": 1.7178, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.7442446947097778, | |
| "rewards/margins": 0.07703767716884613, | |
| "rewards/rejected": -0.8212822675704956, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.17100868403473615, | |
| "grad_norm": 5.480035917246929, | |
| "learning_rate": 4.924223800941717e-07, | |
| "logits/chosen": -0.5283256769180298, | |
| "logits/rejected": -0.5104913711547852, | |
| "logps/chosen": -0.7648738622665405, | |
| "logps/rejected": -0.7856495976448059, | |
| "loss": 1.7199, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.7648738622665405, | |
| "rewards/margins": 0.020775679498910904, | |
| "rewards/rejected": -0.7856495976448059, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.18169672678690715, | |
| "grad_norm": 4.1313808351939345, | |
| "learning_rate": 4.899688103857222e-07, | |
| "logits/chosen": -0.5273383855819702, | |
| "logits/rejected": -0.5292432308197021, | |
| "logps/chosen": -0.7446960210800171, | |
| "logps/rejected": -0.8000040054321289, | |
| "loss": 1.7065, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.7446960210800171, | |
| "rewards/margins": 0.05530792474746704, | |
| "rewards/rejected": -0.8000040054321289, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.19238476953907815, | |
| "grad_norm": 8.284628692622427, | |
| "learning_rate": 4.871796225971999e-07, | |
| "logits/chosen": -0.5416423082351685, | |
| "logits/rejected": -0.5304809808731079, | |
| "logps/chosen": -0.7582653760910034, | |
| "logps/rejected": -0.8476054072380066, | |
| "loss": 1.7129, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.7582653760910034, | |
| "rewards/margins": 0.08933999389410019, | |
| "rewards/rejected": -0.8476054072380066, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.20307281229124916, | |
| "grad_norm": 13.117607752436767, | |
| "learning_rate": 4.840587176599343e-07, | |
| "logits/chosen": -0.4951515793800354, | |
| "logits/rejected": -0.46285510063171387, | |
| "logps/chosen": -0.7567824125289917, | |
| "logps/rejected": -0.7597783803939819, | |
| "loss": 1.7123, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.7567824125289917, | |
| "rewards/margins": 0.0029960013926029205, | |
| "rewards/rejected": -0.7597783803939819, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.21376085504342018, | |
| "grad_norm": 8.62418057833621, | |
| "learning_rate": 4.806104604416823e-07, | |
| "logits/chosen": -0.5064970254898071, | |
| "logits/rejected": -0.475999653339386, | |
| "logps/chosen": -0.7877062559127808, | |
| "logps/rejected": -0.8247254490852356, | |
| "loss": 1.7143, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.7877062559127808, | |
| "rewards/margins": 0.03701917827129364, | |
| "rewards/rejected": -0.8247254490852356, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.22444889779559118, | |
| "grad_norm": 60.910812666712395, | |
| "learning_rate": 4.768396736419662e-07, | |
| "logits/chosen": -0.5239318609237671, | |
| "logits/rejected": -0.5040836334228516, | |
| "logps/chosen": -0.6864551305770874, | |
| "logps/rejected": -0.7309442758560181, | |
| "loss": 1.6974, | |
| "rewards/accuracies": 0.4437499940395355, | |
| "rewards/chosen": -0.6864551305770874, | |
| "rewards/margins": 0.044489093124866486, | |
| "rewards/rejected": -0.7309442758560181, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.23513694054776219, | |
| "grad_norm": 9.487546348283425, | |
| "learning_rate": 4.7275163104709194e-07, | |
| "logits/chosen": -0.5690908432006836, | |
| "logits/rejected": -0.5576282143592834, | |
| "logps/chosen": -0.6891781091690063, | |
| "logps/rejected": -0.7534418106079102, | |
| "loss": 1.6882, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.6891781091690063, | |
| "rewards/margins": 0.06426368653774261, | |
| "rewards/rejected": -0.7534418106079102, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.2458249832999332, | |
| "grad_norm": 12.978514369908993, | |
| "learning_rate": 4.683520501542824e-07, | |
| "logits/chosen": -0.48343902826309204, | |
| "logits/rejected": -0.47186630964279175, | |
| "logps/chosen": -0.7530213594436646, | |
| "logps/rejected": -0.8362549543380737, | |
| "loss": 1.7304, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.7530213594436646, | |
| "rewards/margins": 0.08323363959789276, | |
| "rewards/rejected": -0.8362549543380737, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.2565130260521042, | |
| "grad_norm": 6.953657611353263, | |
| "learning_rate": 4.636470841752404e-07, | |
| "logits/chosen": -0.5461719036102295, | |
| "logits/rejected": -0.5296822786331177, | |
| "logps/chosen": -0.7844793796539307, | |
| "logps/rejected": -0.809302031993866, | |
| "loss": 1.713, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.7844793796539307, | |
| "rewards/margins": 0.024822598323225975, | |
| "rewards/rejected": -0.809302031993866, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.26720106880427524, | |
| "grad_norm": 75.48187194860407, | |
| "learning_rate": 4.5864331343032565e-07, | |
| "logits/chosen": -0.5830751061439514, | |
| "logits/rejected": -0.5749406814575195, | |
| "logps/chosen": -0.727062463760376, | |
| "logps/rejected": -0.8147857785224915, | |
| "loss": 1.6882, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.727062463760376, | |
| "rewards/margins": 0.08772335946559906, | |
| "rewards/rejected": -0.8147857785224915, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.27788911155644624, | |
| "grad_norm": 26.565232629788884, | |
| "learning_rate": 4.533477361453819e-07, | |
| "logits/chosen": -0.4935227036476135, | |
| "logits/rejected": -0.4637245237827301, | |
| "logps/chosen": -0.7516862750053406, | |
| "logps/rejected": -0.7871814966201782, | |
| "loss": 1.7228, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.7516862750053406, | |
| "rewards/margins": 0.035495221614837646, | |
| "rewards/rejected": -0.7871814966201782, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.28857715430861725, | |
| "grad_norm": 6.469835005836861, | |
| "learning_rate": 4.4776775866408533e-07, | |
| "logits/chosen": -0.5782444477081299, | |
| "logits/rejected": -0.5481707453727722, | |
| "logps/chosen": -0.7338271141052246, | |
| "logps/rejected": -0.7812387347221375, | |
| "loss": 1.7057, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.7338271141052246, | |
| "rewards/margins": 0.047411661595106125, | |
| "rewards/rejected": -0.7812387347221375, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.29926519706078825, | |
| "grad_norm": 11.464769788156609, | |
| "learning_rate": 4.4191118508950277e-07, | |
| "logits/chosen": -0.6259862184524536, | |
| "logits/rejected": -0.6170923113822937, | |
| "logps/chosen": -0.7845194339752197, | |
| "logps/rejected": -0.8255494832992554, | |
| "loss": 1.7102, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.7845194339752197, | |
| "rewards/margins": 0.04103006795048714, | |
| "rewards/rejected": -0.8255494832992554, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.30995323981295925, | |
| "grad_norm": 5.222666268344881, | |
| "learning_rate": 4.357862063693485e-07, | |
| "logits/chosen": -0.5420253872871399, | |
| "logits/rejected": -0.5192813873291016, | |
| "logps/chosen": -0.7556418180465698, | |
| "logps/rejected": -0.8247256278991699, | |
| "loss": 1.6915, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.7556418180465698, | |
| "rewards/margins": 0.06908373534679413, | |
| "rewards/rejected": -0.8247256278991699, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.32064128256513025, | |
| "grad_norm": 21.552487521417813, | |
| "learning_rate": 4.294013888402029e-07, | |
| "logits/chosen": -0.5467855334281921, | |
| "logits/rejected": -0.5231701135635376, | |
| "logps/chosen": -0.7533959150314331, | |
| "logps/rejected": -0.8603185415267944, | |
| "loss": 1.6795, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.7533959150314331, | |
| "rewards/margins": 0.10692262649536133, | |
| "rewards/rejected": -0.8603185415267944, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.33132932531730125, | |
| "grad_norm": 9.160529099102835, | |
| "learning_rate": 4.227656622467162e-07, | |
| "logits/chosen": -0.5489504337310791, | |
| "logits/rejected": -0.5380881428718567, | |
| "logps/chosen": -0.7442641854286194, | |
| "logps/rejected": -0.7998737096786499, | |
| "loss": 1.6879, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.7442641854286194, | |
| "rewards/margins": 0.055609602481126785, | |
| "rewards/rejected": -0.7998737096786499, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.3420173680694723, | |
| "grad_norm": 13.838986981177094, | |
| "learning_rate": 4.158883072525528e-07, | |
| "logits/chosen": -0.5547603368759155, | |
| "logits/rejected": -0.5265741348266602, | |
| "logps/chosen": -0.7662423253059387, | |
| "logps/rejected": -0.8465498685836792, | |
| "loss": 1.7164, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.7662423253059387, | |
| "rewards/margins": 0.08030756562948227, | |
| "rewards/rejected": -0.8465498685836792, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3527054108216433, | |
| "grad_norm": 18.503138956112736, | |
| "learning_rate": 4.087789424605447e-07, | |
| "logits/chosen": -0.4897325932979584, | |
| "logits/rejected": -0.48959702253341675, | |
| "logps/chosen": -0.7147817015647888, | |
| "logps/rejected": -0.7714636325836182, | |
| "loss": 1.6904, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.7147817015647888, | |
| "rewards/margins": 0.05668187141418457, | |
| "rewards/rejected": -0.7714636325836182, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.3633934535738143, | |
| "grad_norm": 5.45975525554216, | |
| "learning_rate": 4.0144751096020497e-07, | |
| "logits/chosen": -0.5356153249740601, | |
| "logits/rejected": -0.5175585746765137, | |
| "logps/chosen": -0.7276521921157837, | |
| "logps/rejected": -0.7552744150161743, | |
| "loss": 1.6885, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.7276521921157837, | |
| "rewards/margins": 0.027622133493423462, | |
| "rewards/rejected": -0.7552744150161743, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.3740814963259853, | |
| "grad_norm": 4.322005393303067, | |
| "learning_rate": 3.939042664214184e-07, | |
| "logits/chosen": -0.5734872221946716, | |
| "logits/rejected": -0.5444021224975586, | |
| "logps/chosen": -0.7526987791061401, | |
| "logps/rejected": -0.7955418825149536, | |
| "loss": 1.7322, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.7526987791061401, | |
| "rewards/margins": 0.042843036353588104, | |
| "rewards/rejected": -0.7955418825149536, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.3847695390781563, | |
| "grad_norm": 9.320322046262614, | |
| "learning_rate": 3.8615975875375676e-07, | |
| "logits/chosen": -0.5780426263809204, | |
| "logits/rejected": -0.5400681495666504, | |
| "logps/chosen": -0.7395938634872437, | |
| "logps/rejected": -0.7802700996398926, | |
| "loss": 1.7034, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.7395938634872437, | |
| "rewards/margins": 0.04067610949277878, | |
| "rewards/rejected": -0.7802700996398926, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.3954575818303273, | |
| "grad_norm": 11.90474115158779, | |
| "learning_rate": 3.7822481935147655e-07, | |
| "logits/chosen": -0.5291032195091248, | |
| "logits/rejected": -0.5109629034996033, | |
| "logps/chosen": -0.7308065295219421, | |
| "logps/rejected": -0.8095780611038208, | |
| "loss": 1.7103, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.7308065295219421, | |
| "rewards/margins": 0.07877146452665329, | |
| "rewards/rejected": -0.8095780611038208, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.4061456245824983, | |
| "grad_norm": 7.0679466196186045, | |
| "learning_rate": 3.7011054594483443e-07, | |
| "logits/chosen": -0.5642856359481812, | |
| "logits/rejected": -0.5205564498901367, | |
| "logps/chosen": -0.723100483417511, | |
| "logps/rejected": -0.7782529592514038, | |
| "loss": 1.7016, | |
| "rewards/accuracies": 0.45625001192092896, | |
| "rewards/chosen": -0.723100483417511, | |
| "rewards/margins": 0.055152542889118195, | |
| "rewards/rejected": -0.7782529592514038, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.4168336673346693, | |
| "grad_norm": 7.831846673193328, | |
| "learning_rate": 3.618282870789081e-07, | |
| "logits/chosen": -0.5772170424461365, | |
| "logits/rejected": -0.5436347723007202, | |
| "logps/chosen": -0.7883816957473755, | |
| "logps/rejected": -0.8733705282211304, | |
| "loss": 1.698, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.7883816957473755, | |
| "rewards/margins": 0.08498884737491608, | |
| "rewards/rejected": -0.8733705282211304, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.42752171008684037, | |
| "grad_norm": 10.593421768255501, | |
| "learning_rate": 3.5338962624163016e-07, | |
| "logits/chosen": -0.5145230889320374, | |
| "logits/rejected": -0.5036167502403259, | |
| "logps/chosen": -0.7142393589019775, | |
| "logps/rejected": -0.8125501871109009, | |
| "loss": 1.6867, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.7142393589019775, | |
| "rewards/margins": 0.09831082075834274, | |
| "rewards/rejected": -0.8125501871109009, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.43820975283901137, | |
| "grad_norm": 5.78293219216704, | |
| "learning_rate": 3.448063656632321e-07, | |
| "logits/chosen": -0.5741449594497681, | |
| "logits/rejected": -0.5523878335952759, | |
| "logps/chosen": -0.747484028339386, | |
| "logps/rejected": -0.8042131662368774, | |
| "loss": 1.7024, | |
| "rewards/accuracies": 0.59375, | |
| "rewards/chosen": -0.747484028339386, | |
| "rewards/margins": 0.05672904849052429, | |
| "rewards/rejected": -0.8042131662368774, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.44889779559118237, | |
| "grad_norm": 20.623642957147894, | |
| "learning_rate": 3.360905098097587e-07, | |
| "logits/chosen": -0.6031004190444946, | |
| "logits/rejected": -0.5714393854141235, | |
| "logps/chosen": -0.7718087434768677, | |
| "logps/rejected": -0.8422368168830872, | |
| "loss": 1.703, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.7718087434768677, | |
| "rewards/margins": 0.07042801380157471, | |
| "rewards/rejected": -0.8422368168830872, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.45958583834335337, | |
| "grad_norm": 15.987141543194983, | |
| "learning_rate": 3.272542485937368e-07, | |
| "logits/chosen": -0.5373409986495972, | |
| "logits/rejected": -0.5175820589065552, | |
| "logps/chosen": -0.716375470161438, | |
| "logps/rejected": -0.7990888357162476, | |
| "loss": 1.6872, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.716375470161438, | |
| "rewards/margins": 0.08271351456642151, | |
| "rewards/rejected": -0.7990888357162476, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.47027388109552437, | |
| "grad_norm": 6.205126197800984, | |
| "learning_rate": 3.1830994032548e-07, | |
| "logits/chosen": -0.578917384147644, | |
| "logits/rejected": -0.5596814751625061, | |
| "logps/chosen": -0.7113552689552307, | |
| "logps/rejected": -0.8279391527175903, | |
| "loss": 1.687, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.7113552689552307, | |
| "rewards/margins": 0.1165839210152626, | |
| "rewards/rejected": -0.8279391527175903, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.48096192384769537, | |
| "grad_norm": 10.046269490556881, | |
| "learning_rate": 3.0927009442887437e-07, | |
| "logits/chosen": -0.5646452903747559, | |
| "logits/rejected": -0.5401166677474976, | |
| "logps/chosen": -0.7311114072799683, | |
| "logps/rejected": -0.8085540533065796, | |
| "loss": 1.6805, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.7311114072799683, | |
| "rewards/margins": 0.07744277268648148, | |
| "rewards/rejected": -0.8085540533065796, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.4916499665998664, | |
| "grad_norm": 60.55905374276313, | |
| "learning_rate": 3.001473539458182e-07, | |
| "logits/chosen": -0.5465862154960632, | |
| "logits/rejected": -0.5180550813674927, | |
| "logps/chosen": -0.7326877117156982, | |
| "logps/rejected": -0.8041432499885559, | |
| "loss": 1.6878, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.7326877117156982, | |
| "rewards/margins": 0.07145554572343826, | |
| "rewards/rejected": -0.8041432499885559, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.5023380093520374, | |
| "grad_norm": 19.098707609035294, | |
| "learning_rate": 2.909544778537844e-07, | |
| "logits/chosen": -0.5346301794052124, | |
| "logits/rejected": -0.5442434549331665, | |
| "logps/chosen": -0.7525485157966614, | |
| "logps/rejected": -0.842462420463562, | |
| "loss": 1.6936, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.7525485157966614, | |
| "rewards/margins": 0.08991402387619019, | |
| "rewards/rejected": -0.842462420463562, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.5130260521042084, | |
| "grad_norm": 7.237806739472835, | |
| "learning_rate": 2.817043232212371e-07, | |
| "logits/chosen": -0.5800755620002747, | |
| "logits/rejected": -0.5725291967391968, | |
| "logps/chosen": -0.7127925157546997, | |
| "logps/rejected": -0.8328620195388794, | |
| "loss": 1.7068, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.7127925157546997, | |
| "rewards/margins": 0.12006942927837372, | |
| "rewards/rejected": -0.8328620195388794, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.5237140948563794, | |
| "grad_norm": 17.52718126957788, | |
| "learning_rate": 2.7240982722585837e-07, | |
| "logits/chosen": -0.5736847519874573, | |
| "logits/rejected": -0.5583214163780212, | |
| "logps/chosen": -0.7178992033004761, | |
| "logps/rejected": -0.7496196031570435, | |
| "loss": 1.6766, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.7178992033004761, | |
| "rewards/margins": 0.03172041475772858, | |
| "rewards/rejected": -0.7496196031570435, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.5344021376085505, | |
| "grad_norm": 10.473779064231168, | |
| "learning_rate": 2.63083989060736e-07, | |
| "logits/chosen": -0.5629764795303345, | |
| "logits/rejected": -0.5487209558486938, | |
| "logps/chosen": -0.7848128080368042, | |
| "logps/rejected": -0.8497540354728699, | |
| "loss": 1.6826, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.7848128080368042, | |
| "rewards/margins": 0.06494127959012985, | |
| "rewards/rejected": -0.8497540354728699, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.5450901803607214, | |
| "grad_norm": 12.871682096167538, | |
| "learning_rate": 2.537398517538159e-07, | |
| "logits/chosen": -0.5813488960266113, | |
| "logits/rejected": -0.5667364001274109, | |
| "logps/chosen": -0.7971448302268982, | |
| "logps/rejected": -0.8524147272109985, | |
| "loss": 1.7044, | |
| "rewards/accuracies": 0.45625001192092896, | |
| "rewards/chosen": -0.7971448302268982, | |
| "rewards/margins": 0.055269915610551834, | |
| "rewards/rejected": -0.8524147272109985, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.5557782231128925, | |
| "grad_norm": 50.05319500780107, | |
| "learning_rate": 2.4439048392604877e-07, | |
| "logits/chosen": -0.5521305799484253, | |
| "logits/rejected": -0.5354727506637573, | |
| "logps/chosen": -0.7327480912208557, | |
| "logps/rejected": -0.821591854095459, | |
| "loss": 1.69, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.7327480912208557, | |
| "rewards/margins": 0.08884377777576447, | |
| "rewards/rejected": -0.821591854095459, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.5664662658650634, | |
| "grad_norm": 11.91226212649796, | |
| "learning_rate": 2.3504896151374144e-07, | |
| "logits/chosen": -0.5424870252609253, | |
| "logits/rejected": -0.5289402604103088, | |
| "logps/chosen": -0.6943300366401672, | |
| "logps/rejected": -0.7736053466796875, | |
| "loss": 1.6949, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.6943300366401672, | |
| "rewards/margins": 0.07927526533603668, | |
| "rewards/rejected": -0.7736053466796875, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.5771543086172345, | |
| "grad_norm": 7.15031892491476, | |
| "learning_rate": 2.2572834948067795e-07, | |
| "logits/chosen": -0.5721378326416016, | |
| "logits/rejected": -0.5560299754142761, | |
| "logps/chosen": -0.728114664554596, | |
| "logps/rejected": -0.8583661913871765, | |
| "loss": 1.6823, | |
| "rewards/accuracies": 0.5625, | |
| "rewards/chosen": -0.728114664554596, | |
| "rewards/margins": 0.13025154173374176, | |
| "rewards/rejected": -0.8583661913871765, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.5878423513694054, | |
| "grad_norm": 5.1672000439587835, | |
| "learning_rate": 2.164416835455862e-07, | |
| "logits/chosen": -0.5656030774116516, | |
| "logits/rejected": -0.5321778655052185, | |
| "logps/chosen": -0.7288385033607483, | |
| "logps/rejected": -0.7861572504043579, | |
| "loss": 1.6747, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.7288385033607483, | |
| "rewards/margins": 0.057318806648254395, | |
| "rewards/rejected": -0.7861572504043579, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.5985303941215765, | |
| "grad_norm": 5.10382704888688, | |
| "learning_rate": 2.072019519505062e-07, | |
| "logits/chosen": -0.5811439156532288, | |
| "logits/rejected": -0.5904663801193237, | |
| "logps/chosen": -0.7572720646858215, | |
| "logps/rejected": -0.8319457769393921, | |
| "loss": 1.6921, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.7572720646858215, | |
| "rewards/margins": 0.07467377930879593, | |
| "rewards/rejected": -0.8319457769393921, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.6092184368737475, | |
| "grad_norm": 6.216243824832624, | |
| "learning_rate": 1.980220772955602e-07, | |
| "logits/chosen": -0.5445040464401245, | |
| "logits/rejected": -0.5262423157691956, | |
| "logps/chosen": -0.7155659794807434, | |
| "logps/rejected": -0.7816120386123657, | |
| "loss": 1.6955, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.7155659794807434, | |
| "rewards/margins": 0.06604593247175217, | |
| "rewards/rejected": -0.7816120386123657, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.6199064796259185, | |
| "grad_norm": 11.012279431743695, | |
| "learning_rate": 1.8891489846552644e-07, | |
| "logits/chosen": -0.5420463681221008, | |
| "logits/rejected": -0.5186786651611328, | |
| "logps/chosen": -0.7754839062690735, | |
| "logps/rejected": -0.8797551989555359, | |
| "loss": 1.7328, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.7754839062690735, | |
| "rewards/margins": 0.1042712926864624, | |
| "rewards/rejected": -0.8797551989555359, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.6305945223780896, | |
| "grad_norm": 8.531563186721904, | |
| "learning_rate": 1.7989315267349933e-07, | |
| "logits/chosen": -0.5110348463058472, | |
| "logits/rejected": -0.5081480145454407, | |
| "logps/chosen": -0.6379803419113159, | |
| "logps/rejected": -0.7608265280723572, | |
| "loss": 1.7042, | |
| "rewards/accuracies": 0.6000000238418579, | |
| "rewards/chosen": -0.6379803419113159, | |
| "rewards/margins": 0.122846320271492, | |
| "rewards/rejected": -0.7608265280723572, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.6412825651302605, | |
| "grad_norm": 13.248413116368877, | |
| "learning_rate": 1.7096945764674398e-07, | |
| "logits/chosen": -0.5741692185401917, | |
| "logits/rejected": -0.5570812821388245, | |
| "logps/chosen": -0.7218527793884277, | |
| "logps/rejected": -0.892599880695343, | |
| "loss": 1.6662, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.7218527793884277, | |
| "rewards/margins": 0.17074710130691528, | |
| "rewards/rejected": -0.892599880695343, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.6519706078824316, | |
| "grad_norm": 4.8948399550412995, | |
| "learning_rate": 1.621562939796643e-07, | |
| "logits/chosen": -0.596317708492279, | |
| "logits/rejected": -0.5838618874549866, | |
| "logps/chosen": -0.7185841798782349, | |
| "logps/rejected": -0.8120265007019043, | |
| "loss": 1.6862, | |
| "rewards/accuracies": 0.5874999761581421, | |
| "rewards/chosen": -0.7185841798782349, | |
| "rewards/margins": 0.09344234317541122, | |
| "rewards/rejected": -0.8120265007019043, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.6626586506346025, | |
| "grad_norm": 4.84975331040732, | |
| "learning_rate": 1.5346598767856345e-07, | |
| "logits/chosen": -0.5950718522071838, | |
| "logits/rejected": -0.5880438089370728, | |
| "logps/chosen": -0.6933035254478455, | |
| "logps/rejected": -0.7816718220710754, | |
| "loss": 1.688, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.6933035254478455, | |
| "rewards/margins": 0.08836831152439117, | |
| "rewards/rejected": -0.7816718220710754, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.6733466933867736, | |
| "grad_norm": 103.70714396784868, | |
| "learning_rate": 1.4491069292260866e-07, | |
| "logits/chosen": -0.6049012541770935, | |
| "logits/rejected": -0.5747203826904297, | |
| "logps/chosen": -0.7827696800231934, | |
| "logps/rejected": -0.8818681836128235, | |
| "loss": 1.706, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.7827696800231934, | |
| "rewards/margins": 0.09909845888614655, | |
| "rewards/rejected": -0.8818681836128235, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.6840347361389446, | |
| "grad_norm": 13.54520905911542, | |
| "learning_rate": 1.365023750651133e-07, | |
| "logits/chosen": -0.5884140729904175, | |
| "logits/rejected": -0.5558395385742188, | |
| "logps/chosen": -0.6824565529823303, | |
| "logps/rejected": -0.7401825189590454, | |
| "loss": 1.695, | |
| "rewards/accuracies": 0.5062500238418579, | |
| "rewards/chosen": -0.6824565529823303, | |
| "rewards/margins": 0.05772595480084419, | |
| "rewards/rejected": -0.7401825189590454, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.6947227788911156, | |
| "grad_norm": 12.549862022773366, | |
| "learning_rate": 1.2825279389890818e-07, | |
| "logits/chosen": -0.5767480134963989, | |
| "logits/rejected": -0.5533634424209595, | |
| "logps/chosen": -0.7307712435722351, | |
| "logps/rejected": -0.8965225219726562, | |
| "loss": 1.6925, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.7307712435722351, | |
| "rewards/margins": 0.16575121879577637, | |
| "rewards/rejected": -0.8965225219726562, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.7054108216432866, | |
| "grad_norm": 5.896284097713048, | |
| "learning_rate": 1.201734872092077e-07, | |
| "logits/chosen": -0.5811389684677124, | |
| "logits/rejected": -0.566686749458313, | |
| "logps/chosen": -0.7097315788269043, | |
| "logps/rejected": -0.7522233128547668, | |
| "loss": 1.6769, | |
| "rewards/accuracies": 0.53125, | |
| "rewards/chosen": -0.7097315788269043, | |
| "rewards/margins": 0.04249165579676628, | |
| "rewards/rejected": -0.7522233128547668, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.7160988643954576, | |
| "grad_norm": 6.411999544175617, | |
| "learning_rate": 1.1227575463697439e-07, | |
| "logits/chosen": -0.5806372761726379, | |
| "logits/rejected": -0.5382911562919617, | |
| "logps/chosen": -0.7409597635269165, | |
| "logps/rejected": -0.8579269647598267, | |
| "loss": 1.71, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.7409597635269165, | |
| "rewards/margins": 0.11696724593639374, | |
| "rewards/rejected": -0.8579269647598267, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.7267869071476286, | |
| "grad_norm": 14.609165506525063, | |
| "learning_rate": 1.0457064187534861e-07, | |
| "logits/chosen": -0.5816670656204224, | |
| "logits/rejected": -0.5706329345703125, | |
| "logps/chosen": -0.7048308849334717, | |
| "logps/rejected": -0.831493079662323, | |
| "loss": 1.7101, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.7048308849334717, | |
| "rewards/margins": 0.1266622394323349, | |
| "rewards/rejected": -0.831493079662323, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.7374749498997996, | |
| "grad_norm": 7.153574940132466, | |
| "learning_rate": 9.706892522124838e-08, | |
| "logits/chosen": -0.6188436150550842, | |
| "logits/rejected": -0.6014577150344849, | |
| "logps/chosen": -0.7070926427841187, | |
| "logps/rejected": -0.7998514771461487, | |
| "loss": 1.6995, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.7070926427841187, | |
| "rewards/margins": 0.09275877475738525, | |
| "rewards/rejected": -0.7998514771461487, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.7481629926519706, | |
| "grad_norm": 5.896246793242024, | |
| "learning_rate": 8.978109650374396e-08, | |
| "logits/chosen": -0.5400093793869019, | |
| "logits/rejected": -0.5397418737411499, | |
| "logps/chosen": -0.7316173315048218, | |
| "logps/rejected": -0.8627879023551941, | |
| "loss": 1.6711, | |
| "rewards/accuracies": 0.574999988079071, | |
| "rewards/chosen": -0.7316173315048218, | |
| "rewards/margins": 0.13117054104804993, | |
| "rewards/rejected": -0.8627879023551941, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7588510354041417, | |
| "grad_norm": 5.038798783618942, | |
| "learning_rate": 8.271734841028552e-08, | |
| "logits/chosen": -0.5678108930587769, | |
| "logits/rejected": -0.5479924082756042, | |
| "logps/chosen": -0.7514128088951111, | |
| "logps/rejected": -0.9308807253837585, | |
| "loss": 1.6837, | |
| "rewards/accuracies": 0.550000011920929, | |
| "rewards/chosen": -0.7514128088951111, | |
| "rewards/margins": 0.17946788668632507, | |
| "rewards/rejected": -0.9308807253837585, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.7695390781563126, | |
| "grad_norm": 13.150430860121316, | |
| "learning_rate": 7.588756023130833e-08, | |
| "logits/chosen": -0.5509608387947083, | |
| "logits/rejected": -0.5354821681976318, | |
| "logps/chosen": -0.7503410577774048, | |
| "logps/rejected": -0.8366758227348328, | |
| "loss": 1.7028, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.7503410577774048, | |
| "rewards/margins": 0.08633477985858917, | |
| "rewards/rejected": -0.8366758227348328, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.7802271209084837, | |
| "grad_norm": 6.429127672704049, | |
| "learning_rate": 6.930128404315214e-08, | |
| "logits/chosen": -0.6169145703315735, | |
| "logits/rejected": -0.616595983505249, | |
| "logps/chosen": -0.7533454895019531, | |
| "logps/rejected": -0.8524212837219238, | |
| "loss": 1.7026, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.7533454895019531, | |
| "rewards/margins": 0.09907577186822891, | |
| "rewards/rejected": -0.8524212837219238, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.7909151636606546, | |
| "grad_norm": 6.289464178098906, | |
| "learning_rate": 6.296773134861824e-08, | |
| "logits/chosen": -0.546366810798645, | |
| "logits/rejected": -0.5380454659461975, | |
| "logps/chosen": -0.7598742246627808, | |
| "logps/rejected": -0.811627984046936, | |
| "loss": 1.6928, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.7598742246627808, | |
| "rewards/margins": 0.051753778010606766, | |
| "rewards/rejected": -0.811627984046936, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.8016032064128257, | |
| "grad_norm": 5.398411336966849, | |
| "learning_rate": 5.6895760193850145e-08, | |
| "logits/chosen": -0.5729818344116211, | |
| "logits/rejected": -0.5412513017654419, | |
| "logps/chosen": -0.7241582870483398, | |
| "logps/rejected": -0.8142971992492676, | |
| "loss": 1.6808, | |
| "rewards/accuracies": 0.543749988079071, | |
| "rewards/chosen": -0.7241582870483398, | |
| "rewards/margins": 0.09013892710208893, | |
| "rewards/rejected": -0.8142971992492676, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.8122912491649966, | |
| "grad_norm": 9.274040284504924, | |
| "learning_rate": 5.109386277955477e-08, | |
| "logits/chosen": -0.5896815061569214, | |
| "logits/rejected": -0.5692397356033325, | |
| "logps/chosen": -0.7495406270027161, | |
| "logps/rejected": -0.9361629486083984, | |
| "loss": 1.6883, | |
| "rewards/accuracies": 0.581250011920929, | |
| "rewards/chosen": -0.7495406270027161, | |
| "rewards/margins": 0.1866222769021988, | |
| "rewards/rejected": -0.9361629486083984, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.8229792919171677, | |
| "grad_norm": 9.755234176801242, | |
| "learning_rate": 4.557015358389216e-08, | |
| "logits/chosen": -0.5829408764839172, | |
| "logits/rejected": -0.5578514337539673, | |
| "logps/chosen": -0.7497562766075134, | |
| "logps/rejected": -0.8292330503463745, | |
| "loss": 1.6962, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.7497562766075134, | |
| "rewards/margins": 0.07947687804698944, | |
| "rewards/rejected": -0.8292330503463745, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.8336673346693386, | |
| "grad_norm": 13.589407312863031, | |
| "learning_rate": 4.0332358013644015e-08, | |
| "logits/chosen": -0.5740140676498413, | |
| "logits/rejected": -0.5624168515205383, | |
| "logps/chosen": -0.7605090141296387, | |
| "logps/rejected": -0.8072217106819153, | |
| "loss": 1.6945, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.7605090141296387, | |
| "rewards/margins": 0.046712733805179596, | |
| "rewards/rejected": -0.8072217106819153, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.8443553774215097, | |
| "grad_norm": 5.944832078088462, | |
| "learning_rate": 3.538780159953347e-08, | |
| "logits/chosen": -0.5879320502281189, | |
| "logits/rejected": -0.5540255308151245, | |
| "logps/chosen": -0.7190330028533936, | |
| "logps/rejected": -0.7648254632949829, | |
| "loss": 1.7036, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.7190330028533936, | |
| "rewards/margins": 0.04579242318868637, | |
| "rewards/rejected": -0.7648254632949829, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.8550434201736807, | |
| "grad_norm": 7.162063518571142, | |
| "learning_rate": 3.074339975080836e-08, | |
| "logits/chosen": -0.5846894979476929, | |
| "logits/rejected": -0.5598914623260498, | |
| "logps/chosen": -0.7635160684585571, | |
| "logps/rejected": -0.835554301738739, | |
| "loss": 1.6906, | |
| "rewards/accuracies": 0.48750001192092896, | |
| "rewards/chosen": -0.7635160684585571, | |
| "rewards/margins": 0.07203832268714905, | |
| "rewards/rejected": -0.835554301738739, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8550434201736807, | |
| "eval_logits/chosen": -0.5922300815582275, | |
| "eval_logits/rejected": -0.5793322920799255, | |
| "eval_logps/chosen": -0.7343553900718689, | |
| "eval_logps/rejected": -0.8094466328620911, | |
| "eval_loss": 1.6913269758224487, | |
| "eval_rewards/accuracies": 0.5609756112098694, | |
| "eval_rewards/chosen": -0.7343553900718689, | |
| "eval_rewards/margins": 0.07509122788906097, | |
| "eval_rewards/rejected": -0.8094466328620911, | |
| "eval_runtime": 427.2359, | |
| "eval_samples_per_second": 4.59, | |
| "eval_steps_per_second": 0.288, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.8657314629258517, | |
| "grad_norm": 9.600105525125768, | |
| "learning_rate": 2.6405648083415833e-08, | |
| "logits/chosen": -0.6273748874664307, | |
| "logits/rejected": -0.598314642906189, | |
| "logps/chosen": -0.7624421119689941, | |
| "logps/rejected": -0.7932037115097046, | |
| "loss": 1.6796, | |
| "rewards/accuracies": 0.46875, | |
| "rewards/chosen": -0.7624421119689941, | |
| "rewards/margins": 0.030761664733290672, | |
| "rewards/rejected": -0.7932037115097046, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 0.8764195056780227, | |
| "grad_norm": 7.038480876123261, | |
| "learning_rate": 2.2380613335296033e-08, | |
| "logits/chosen": -0.5409439206123352, | |
| "logits/rejected": -0.5284377336502075, | |
| "logps/chosen": -0.7356694340705872, | |
| "logps/rejected": -0.8067193031311035, | |
| "loss": 1.6923, | |
| "rewards/accuracies": 0.518750011920929, | |
| "rewards/chosen": -0.7356694340705872, | |
| "rewards/margins": 0.07104983180761337, | |
| "rewards/rejected": -0.8067193031311035, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.8871075484301937, | |
| "grad_norm": 16.739588675046768, | |
| "learning_rate": 1.8673924881500823e-08, | |
| "logits/chosen": -0.5183539390563965, | |
| "logits/rejected": -0.545289933681488, | |
| "logps/chosen": -0.6668271422386169, | |
| "logps/rejected": -0.7874898314476013, | |
| "loss": 1.6966, | |
| "rewards/accuracies": 0.612500011920929, | |
| "rewards/chosen": -0.6668271422386169, | |
| "rewards/margins": 0.12066280841827393, | |
| "rewards/rejected": -0.7874898314476013, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 0.8977955911823647, | |
| "grad_norm": 5.059249540924062, | |
| "learning_rate": 1.5290766861003475e-08, | |
| "logits/chosen": -0.5611199736595154, | |
| "logits/rejected": -0.5376091599464417, | |
| "logps/chosen": -0.7408109903335571, | |
| "logps/rejected": -0.9328848123550415, | |
| "loss": 1.6789, | |
| "rewards/accuracies": 0.5687500238418579, | |
| "rewards/chosen": -0.7408109903335571, | |
| "rewards/margins": 0.19207359850406647, | |
| "rewards/rejected": -0.9328848123550415, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.9084836339345357, | |
| "grad_norm": 20.81792424320682, | |
| "learning_rate": 1.2235870926211616e-08, | |
| "logits/chosen": -0.5288859605789185, | |
| "logits/rejected": -0.5023664236068726, | |
| "logps/chosen": -0.7220520377159119, | |
| "logps/rejected": -0.8065603971481323, | |
| "loss": 1.6789, | |
| "rewards/accuracies": 0.4937500059604645, | |
| "rewards/chosen": -0.7220520377159119, | |
| "rewards/margins": 0.08450835198163986, | |
| "rewards/rejected": -0.8065603971481323, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.9191716766867067, | |
| "grad_norm": 5.479591554243054, | |
| "learning_rate": 9.513509625323518e-09, | |
| "logits/chosen": -0.5215033888816833, | |
| "logits/rejected": -0.48939695954322815, | |
| "logps/chosen": -0.6924680471420288, | |
| "logps/rejected": -0.7432147264480591, | |
| "loss": 1.6704, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.6924680471420288, | |
| "rewards/margins": 0.0507466085255146, | |
| "rewards/rejected": -0.7432147264480591, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.9298597194388778, | |
| "grad_norm": 10.936235980167218, | |
| "learning_rate": 7.127490426783123e-09, | |
| "logits/chosen": -0.5550135970115662, | |
| "logits/rejected": -0.536861777305603, | |
| "logps/chosen": -0.7217072248458862, | |
| "logps/rejected": -0.8055909872055054, | |
| "loss": 1.6964, | |
| "rewards/accuracies": 0.48124998807907104, | |
| "rewards/chosen": -0.7217072248458862, | |
| "rewards/margins": 0.08388371020555496, | |
| "rewards/rejected": -0.8055909872055054, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 0.9405477621910487, | |
| "grad_norm": 19.094745466245598, | |
| "learning_rate": 5.08115039419113e-09, | |
| "logits/chosen": -0.5761109590530396, | |
| "logits/rejected": -0.5287821292877197, | |
| "logps/chosen": -0.7307227849960327, | |
| "logps/rejected": -0.7685351371765137, | |
| "loss": 1.6952, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.7307227849960327, | |
| "rewards/margins": 0.037812262773513794, | |
| "rewards/rejected": -0.7685351371765137, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.9512358049432198, | |
| "grad_norm": 7.894577052872913, | |
| "learning_rate": 3.3773515191196646e-09, | |
| "logits/chosen": -0.5409837961196899, | |
| "logits/rejected": -0.5275704264640808, | |
| "logps/chosen": -0.727105438709259, | |
| "logps/rejected": -0.8007614016532898, | |
| "loss": 1.7049, | |
| "rewards/accuracies": 0.5249999761581421, | |
| "rewards/chosen": -0.727105438709259, | |
| "rewards/margins": 0.07365605980157852, | |
| "rewards/rejected": -0.8007614016532898, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 0.9619238476953907, | |
| "grad_norm": 5.5185430682531065, | |
| "learning_rate": 2.0184767183584474e-09, | |
| "logits/chosen": -0.5929439067840576, | |
| "logits/rejected": -0.5756229162216187, | |
| "logps/chosen": -0.7496530413627625, | |
| "logps/rejected": -0.8659397959709167, | |
| "loss": 1.6809, | |
| "rewards/accuracies": 0.512499988079071, | |
| "rewards/chosen": -0.7496530413627625, | |
| "rewards/margins": 0.11628679931163788, | |
| "rewards/rejected": -0.8659397959709167, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.9726118904475618, | |
| "grad_norm": 5.664821209840954, | |
| "learning_rate": 1.0064265011902328e-09, | |
| "logits/chosen": -0.5561838746070862, | |
| "logits/rejected": -0.5247195959091187, | |
| "logps/chosen": -0.7479305267333984, | |
| "logps/rejected": -0.8220788240432739, | |
| "loss": 1.7264, | |
| "rewards/accuracies": 0.5375000238418579, | |
| "rewards/chosen": -0.7479305267333984, | |
| "rewards/margins": 0.07414824515581131, | |
| "rewards/rejected": -0.8220788240432739, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 0.9832999331997327, | |
| "grad_norm": 5.565216047744654, | |
| "learning_rate": 3.4261631135654167e-10, | |
| "logits/chosen": -0.5442866086959839, | |
| "logits/rejected": -0.5325323939323425, | |
| "logps/chosen": -0.7228007912635803, | |
| "logps/rejected": -0.769408643245697, | |
| "loss": 1.7012, | |
| "rewards/accuracies": 0.5, | |
| "rewards/chosen": -0.7228007912635803, | |
| "rewards/margins": 0.04660790413618088, | |
| "rewards/rejected": -0.769408643245697, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.9939879759519038, | |
| "grad_norm": 8.061696670344013, | |
| "learning_rate": 2.797454743164174e-11, | |
| "logits/chosen": -0.567788302898407, | |
| "logits/rejected": -0.5469498634338379, | |
| "logps/chosen": -0.7121998071670532, | |
| "logps/rejected": -0.8072344660758972, | |
| "loss": 1.6901, | |
| "rewards/accuracies": 0.5562499761581421, | |
| "rewards/chosen": -0.7121998071670532, | |
| "rewards/margins": 0.09503473341464996, | |
| "rewards/rejected": -0.8072344660758972, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 0.9982631930527722, | |
| "step": 467, | |
| "total_flos": 0.0, | |
| "train_loss": 1.7001326604879898, | |
| "train_runtime": 20882.5667, | |
| "train_samples_per_second": 2.867, | |
| "train_steps_per_second": 0.022 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 467, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |