{ "best_global_step": 11236, "best_metric": 0.5682024245488293, "best_model_checkpoint": "runs/clip_mtl_mmhshateful\\checkpoint-11236", "epoch": 1.0, "eval_steps": 500, "global_step": 11236, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.0044499822000712, "grad_norm": 0.6930296421051025, "learning_rate": 1.9999985928231315e-05, "loss": 1.1543, "step": 50 }, { "epoch": 0.0088999644001424, "grad_norm": 0.9534976482391357, "learning_rate": 1.99999413432439e-05, "loss": 1.0927, "step": 100 }, { "epoch": 0.013349946600213599, "grad_norm": 0.8714975118637085, "learning_rate": 1.9999866220743754e-05, "loss": 0.9653, "step": 150 }, { "epoch": 0.0177999288002848, "grad_norm": 1.0742043256759644, "learning_rate": 1.9999760560960283e-05, "loss": 1.2235, "step": 200 }, { "epoch": 0.022249911000355997, "grad_norm": 2.212294101715088, "learning_rate": 1.9999624364216147e-05, "loss": 1.371, "step": 250 }, { "epoch": 0.026699893200427198, "grad_norm": 3.4641246795654297, "learning_rate": 1.999946488655439e-05, "loss": 1.205, "step": 300 }, { "epoch": 0.0311498754004984, "grad_norm": 1.0936462879180908, "learning_rate": 1.9999268838660392e-05, "loss": 0.9664, "step": 350 }, { "epoch": 0.0355998576005696, "grad_norm": 1.8520387411117554, "learning_rate": 1.9999042255307332e-05, "loss": 1.0667, "step": 400 }, { "epoch": 0.0400498398006408, "grad_norm": 1.1343063116073608, "learning_rate": 1.999878513718715e-05, "loss": 1.0061, "step": 450 }, { "epoch": 0.044499822000711994, "grad_norm": 1.8512768745422363, "learning_rate": 1.9998497485085022e-05, "loss": 1.1225, "step": 500 }, { "epoch": 0.048949804200783195, "grad_norm": 11.832172393798828, "learning_rate": 1.999817929987937e-05, "loss": 1.1131, "step": 550 }, { "epoch": 0.053399786400854396, "grad_norm": 11.022147178649902, "learning_rate": 1.9997830582541863e-05, "loss": 1.0418, "step": 600 }, { "epoch": 0.0578497686009256, "grad_norm": 1.15188729763031, "learning_rate": 1.9997459218302607e-05, "loss": 1.384, "step": 650 }, { "epoch": 0.0622997508009968, "grad_norm": 0.8350585699081421, "learning_rate": 1.9997058533115497e-05, "loss": 1.0089, "step": 700 }, { "epoch": 0.066749733001068, "grad_norm": 2.022878885269165, "learning_rate": 1.999661944726586e-05, "loss": 1.0024, "step": 750 }, { "epoch": 0.0711997152011392, "grad_norm": 1.1452044248580933, "learning_rate": 1.9996149834047797e-05, "loss": 1.1402, "step": 800 }, { "epoch": 0.0756496974012104, "grad_norm": 0.855829119682312, "learning_rate": 1.9995649694895397e-05, "loss": 0.9612, "step": 850 }, { "epoch": 0.0800996796012816, "grad_norm": 11.837068557739258, "learning_rate": 1.9995119031335967e-05, "loss": 1.0198, "step": 900 }, { "epoch": 0.08454966180135279, "grad_norm": 0.9849753379821777, "learning_rate": 1.999455784499003e-05, "loss": 1.2736, "step": 950 }, { "epoch": 0.08899964400142399, "grad_norm": 1.7843427658081055, "learning_rate": 1.9993966137571323e-05, "loss": 1.0706, "step": 1000 }, { "epoch": 0.09344962620149519, "grad_norm": 3.2881357669830322, "learning_rate": 1.999334391088678e-05, "loss": 1.1104, "step": 1050 }, { "epoch": 0.09789960840156639, "grad_norm": 19.132781982421875, "learning_rate": 1.999269116683654e-05, "loss": 1.1803, "step": 1100 }, { "epoch": 0.10234959060163759, "grad_norm": 1.0581350326538086, "learning_rate": 1.9992007907413935e-05, "loss": 1.1744, "step": 1150 }, { "epoch": 0.10679957280170879, "grad_norm": 3.8312244415283203, "learning_rate": 1.999129413470548e-05, "loss": 1.0072, "step": 1200 }, { "epoch": 0.11124955500177999, "grad_norm": 0.6719951033592224, "learning_rate": 1.999054985089087e-05, "loss": 1.0091, "step": 1250 }, { "epoch": 0.1156995372018512, "grad_norm": 3.9404165744781494, "learning_rate": 1.9989775058242976e-05, "loss": 1.3107, "step": 1300 }, { "epoch": 0.1201495194019224, "grad_norm": 16.099702835083008, "learning_rate": 1.9988969759127837e-05, "loss": 1.2654, "step": 1350 }, { "epoch": 0.1245995016019936, "grad_norm": 0.6801987886428833, "learning_rate": 1.9988133956004657e-05, "loss": 1.1379, "step": 1400 }, { "epoch": 0.1290494838020648, "grad_norm": 0.9160261154174805, "learning_rate": 1.9987267651425776e-05, "loss": 0.9095, "step": 1450 }, { "epoch": 0.133499466002136, "grad_norm": 1.217660665512085, "learning_rate": 1.9986370848036688e-05, "loss": 1.3063, "step": 1500 }, { "epoch": 0.1379494482022072, "grad_norm": 0.7348505258560181, "learning_rate": 1.9985443548576034e-05, "loss": 1.062, "step": 1550 }, { "epoch": 0.1423994304022784, "grad_norm": 12.344593048095703, "learning_rate": 1.9984485755875563e-05, "loss": 1.249, "step": 1600 }, { "epoch": 0.1468494126023496, "grad_norm": 3.754258155822754, "learning_rate": 1.9983497472860156e-05, "loss": 1.1423, "step": 1650 }, { "epoch": 0.1512993948024208, "grad_norm": 1.3361784219741821, "learning_rate": 1.9982478702547802e-05, "loss": 1.1729, "step": 1700 }, { "epoch": 0.155749377002492, "grad_norm": 1.674297571182251, "learning_rate": 1.9981429448049584e-05, "loss": 1.0536, "step": 1750 }, { "epoch": 0.1601993592025632, "grad_norm": 1.6976128816604614, "learning_rate": 1.9980349712569696e-05, "loss": 1.0989, "step": 1800 }, { "epoch": 0.1646493414026344, "grad_norm": 12.235713005065918, "learning_rate": 1.997923949940539e-05, "loss": 1.0191, "step": 1850 }, { "epoch": 0.16909932360270558, "grad_norm": 8.77859115600586, "learning_rate": 1.9978098811947005e-05, "loss": 1.2005, "step": 1900 }, { "epoch": 0.17354930580277678, "grad_norm": 1.0547969341278076, "learning_rate": 1.9976927653677937e-05, "loss": 1.1798, "step": 1950 }, { "epoch": 0.17799928800284798, "grad_norm": 16.349029541015625, "learning_rate": 1.997572602817463e-05, "loss": 1.201, "step": 2000 }, { "epoch": 0.18244927020291918, "grad_norm": 11.384589195251465, "learning_rate": 1.9974493939106575e-05, "loss": 1.0481, "step": 2050 }, { "epoch": 0.18689925240299038, "grad_norm": 10.341683387756348, "learning_rate": 1.9973231390236287e-05, "loss": 1.2775, "step": 2100 }, { "epoch": 0.19134923460306158, "grad_norm": 0.7568284273147583, "learning_rate": 1.9971938385419298e-05, "loss": 0.9456, "step": 2150 }, { "epoch": 0.19579921680313278, "grad_norm": 0.719429075717926, "learning_rate": 1.9970614928604144e-05, "loss": 1.198, "step": 2200 }, { "epoch": 0.20024919900320398, "grad_norm": 11.964608192443848, "learning_rate": 1.996926102383236e-05, "loss": 1.2098, "step": 2250 }, { "epoch": 0.20469918120327518, "grad_norm": 32.06000518798828, "learning_rate": 1.996787667523846e-05, "loss": 1.3231, "step": 2300 }, { "epoch": 0.20914916340334638, "grad_norm": 4.2478108406066895, "learning_rate": 1.996646188704992e-05, "loss": 1.0375, "step": 2350 }, { "epoch": 0.21359914560341758, "grad_norm": 3.896465301513672, "learning_rate": 1.996501666358718e-05, "loss": 1.0432, "step": 2400 }, { "epoch": 0.21804912780348878, "grad_norm": 3.1395533084869385, "learning_rate": 1.9963541009263615e-05, "loss": 1.1409, "step": 2450 }, { "epoch": 0.22249911000355999, "grad_norm": 1.3287521600723267, "learning_rate": 1.996203492858554e-05, "loss": 0.9895, "step": 2500 }, { "epoch": 0.2269490922036312, "grad_norm": 1.846510648727417, "learning_rate": 1.996049842615217e-05, "loss": 1.1018, "step": 2550 }, { "epoch": 0.2313990744037024, "grad_norm": 0.8593488931655884, "learning_rate": 1.9958931506655635e-05, "loss": 1.2726, "step": 2600 }, { "epoch": 0.2358490566037736, "grad_norm": 1.4827325344085693, "learning_rate": 1.995733417488094e-05, "loss": 0.8561, "step": 2650 }, { "epoch": 0.2402990388038448, "grad_norm": 0.7320635318756104, "learning_rate": 1.995570643570597e-05, "loss": 1.0547, "step": 2700 }, { "epoch": 0.244749021003916, "grad_norm": 2.179614305496216, "learning_rate": 1.995404829410147e-05, "loss": 1.2316, "step": 2750 }, { "epoch": 0.2491990032039872, "grad_norm": 3.888511896133423, "learning_rate": 1.995235975513101e-05, "loss": 1.3863, "step": 2800 }, { "epoch": 0.2536489854040584, "grad_norm": 11.913590431213379, "learning_rate": 1.9950640823951013e-05, "loss": 1.0316, "step": 2850 }, { "epoch": 0.2580989676041296, "grad_norm": 1.4524823427200317, "learning_rate": 1.9948891505810685e-05, "loss": 1.1805, "step": 2900 }, { "epoch": 0.2625489498042008, "grad_norm": 0.8905289173126221, "learning_rate": 1.9947111806052044e-05, "loss": 1.2089, "step": 2950 }, { "epoch": 0.266998932004272, "grad_norm": 0.6847230792045593, "learning_rate": 1.9945301730109884e-05, "loss": 1.0075, "step": 3000 }, { "epoch": 0.2714489142043432, "grad_norm": 0.8868160843849182, "learning_rate": 1.9943461283511763e-05, "loss": 0.8334, "step": 3050 }, { "epoch": 0.2758988964044144, "grad_norm": 1.7983331680297852, "learning_rate": 1.994159047187797e-05, "loss": 1.1179, "step": 3100 }, { "epoch": 0.2803488786044856, "grad_norm": 0.7640360593795776, "learning_rate": 1.9939689300921538e-05, "loss": 1.0782, "step": 3150 }, { "epoch": 0.2847988608045568, "grad_norm": 2.1053290367126465, "learning_rate": 1.9937757776448204e-05, "loss": 1.3257, "step": 3200 }, { "epoch": 0.289248843004628, "grad_norm": 2.609022855758667, "learning_rate": 1.99357959043564e-05, "loss": 1.1516, "step": 3250 }, { "epoch": 0.2936988252046992, "grad_norm": 1.6748881340026855, "learning_rate": 1.9933803690637225e-05, "loss": 1.1508, "step": 3300 }, { "epoch": 0.2981488074047704, "grad_norm": 5.622939109802246, "learning_rate": 1.9931781141374447e-05, "loss": 0.8668, "step": 3350 }, { "epoch": 0.3025987896048416, "grad_norm": 2.957125186920166, "learning_rate": 1.992972826274446e-05, "loss": 1.374, "step": 3400 }, { "epoch": 0.3070487718049128, "grad_norm": 1.5037779808044434, "learning_rate": 1.9927645061016278e-05, "loss": 1.1594, "step": 3450 }, { "epoch": 0.311498754004984, "grad_norm": 1.0722275972366333, "learning_rate": 1.9925531542551524e-05, "loss": 1.2182, "step": 3500 }, { "epoch": 0.3159487362050552, "grad_norm": 0.5968232750892639, "learning_rate": 1.9923387713804393e-05, "loss": 1.2872, "step": 3550 }, { "epoch": 0.3203987184051264, "grad_norm": 1.0795742273330688, "learning_rate": 1.9921213581321646e-05, "loss": 1.1279, "step": 3600 }, { "epoch": 0.3248487006051976, "grad_norm": 0.9048451781272888, "learning_rate": 1.991900915174258e-05, "loss": 1.212, "step": 3650 }, { "epoch": 0.3292986828052688, "grad_norm": 9.705388069152832, "learning_rate": 1.9916774431799017e-05, "loss": 1.3017, "step": 3700 }, { "epoch": 0.33374866500534, "grad_norm": 1.7905807495117188, "learning_rate": 1.9914509428315266e-05, "loss": 0.9617, "step": 3750 }, { "epoch": 0.33819864720541115, "grad_norm": 7.1238603591918945, "learning_rate": 1.9912214148208137e-05, "loss": 1.1694, "step": 3800 }, { "epoch": 0.34264862940548235, "grad_norm": 1.6141045093536377, "learning_rate": 1.9909888598486878e-05, "loss": 1.0389, "step": 3850 }, { "epoch": 0.34709861160555355, "grad_norm": 2.56583309173584, "learning_rate": 1.9907532786253183e-05, "loss": 1.2183, "step": 3900 }, { "epoch": 0.35154859380562475, "grad_norm": 1.3100908994674683, "learning_rate": 1.990514671870116e-05, "loss": 1.2688, "step": 3950 }, { "epoch": 0.35599857600569595, "grad_norm": 2.074667453765869, "learning_rate": 1.990273040311731e-05, "loss": 1.091, "step": 4000 }, { "epoch": 0.36044855820576716, "grad_norm": 0.7037174105644226, "learning_rate": 1.9900283846880498e-05, "loss": 0.998, "step": 4050 }, { "epoch": 0.36489854040583836, "grad_norm": 2.9290120601654053, "learning_rate": 1.9897807057461946e-05, "loss": 0.9211, "step": 4100 }, { "epoch": 0.36934852260590956, "grad_norm": 8.646696090698242, "learning_rate": 1.9895300042425196e-05, "loss": 1.2834, "step": 4150 }, { "epoch": 0.37379850480598076, "grad_norm": 0.8763217329978943, "learning_rate": 1.989276280942609e-05, "loss": 1.0883, "step": 4200 }, { "epoch": 0.37824848700605196, "grad_norm": 0.8757887482643127, "learning_rate": 1.989019536621276e-05, "loss": 1.3121, "step": 4250 }, { "epoch": 0.38269846920612316, "grad_norm": 20.988733291625977, "learning_rate": 1.988759772062558e-05, "loss": 1.1183, "step": 4300 }, { "epoch": 0.38714845140619436, "grad_norm": 0.7720549702644348, "learning_rate": 1.9884969880597162e-05, "loss": 1.2385, "step": 4350 }, { "epoch": 0.39159843360626556, "grad_norm": 2.226048469543457, "learning_rate": 1.9882311854152318e-05, "loss": 1.2543, "step": 4400 }, { "epoch": 0.39604841580633676, "grad_norm": 1.131201148033142, "learning_rate": 1.987962364940805e-05, "loss": 1.0842, "step": 4450 }, { "epoch": 0.40049839800640796, "grad_norm": 14.457625389099121, "learning_rate": 1.987690527457352e-05, "loss": 1.0695, "step": 4500 }, { "epoch": 0.40494838020647916, "grad_norm": 0.8496118187904358, "learning_rate": 1.9874156737950008e-05, "loss": 1.2233, "step": 4550 }, { "epoch": 0.40939836240655036, "grad_norm": 1.1709364652633667, "learning_rate": 1.9871378047930915e-05, "loss": 1.0452, "step": 4600 }, { "epoch": 0.41384834460662157, "grad_norm": 0.7952430248260498, "learning_rate": 1.9868569213001722e-05, "loss": 1.1319, "step": 4650 }, { "epoch": 0.41829832680669277, "grad_norm": 10.694504737854004, "learning_rate": 1.9865730241739958e-05, "loss": 1.0488, "step": 4700 }, { "epoch": 0.42274830900676397, "grad_norm": 0.7666745781898499, "learning_rate": 1.986286114281519e-05, "loss": 1.2433, "step": 4750 }, { "epoch": 0.42719829120683517, "grad_norm": 0.5612050890922546, "learning_rate": 1.9859961924988977e-05, "loss": 0.7798, "step": 4800 }, { "epoch": 0.43164827340690637, "grad_norm": 1.1437429189682007, "learning_rate": 1.985703259711487e-05, "loss": 1.0582, "step": 4850 }, { "epoch": 0.43609825560697757, "grad_norm": 1.3851732015609741, "learning_rate": 1.9854073168138354e-05, "loss": 1.2503, "step": 4900 }, { "epoch": 0.44054823780704877, "grad_norm": 0.5759838819503784, "learning_rate": 1.985108364709684e-05, "loss": 1.0713, "step": 4950 }, { "epoch": 0.44499822000711997, "grad_norm": 0.9803907871246338, "learning_rate": 1.9848064043119644e-05, "loss": 1.0747, "step": 5000 }, { "epoch": 0.44944820220719117, "grad_norm": 1.4427608251571655, "learning_rate": 1.9845014365427936e-05, "loss": 0.9888, "step": 5050 }, { "epoch": 0.4538981844072624, "grad_norm": 0.6690992712974548, "learning_rate": 1.984193462333472e-05, "loss": 1.1711, "step": 5100 }, { "epoch": 0.4583481666073336, "grad_norm": 12.372550964355469, "learning_rate": 1.983882482624483e-05, "loss": 1.0621, "step": 5150 }, { "epoch": 0.4627981488074048, "grad_norm": 1.351020336151123, "learning_rate": 1.983568498365486e-05, "loss": 0.9669, "step": 5200 }, { "epoch": 0.467248131007476, "grad_norm": 1.8895344734191895, "learning_rate": 1.9832515105153165e-05, "loss": 1.334, "step": 5250 }, { "epoch": 0.4716981132075472, "grad_norm": 2.086756467819214, "learning_rate": 1.9829315200419824e-05, "loss": 1.1035, "step": 5300 }, { "epoch": 0.4761480954076184, "grad_norm": 2.531623363494873, "learning_rate": 1.9826085279226612e-05, "loss": 0.9576, "step": 5350 }, { "epoch": 0.4805980776076896, "grad_norm": 13.519452095031738, "learning_rate": 1.9822825351436952e-05, "loss": 1.1265, "step": 5400 }, { "epoch": 0.4850480598077608, "grad_norm": Infinity, "learning_rate": 1.9819535427005925e-05, "loss": 0.987, "step": 5450 }, { "epoch": 0.489498042007832, "grad_norm": 2.1377389430999756, "learning_rate": 1.981628220800391e-05, "loss": 1.1994, "step": 5500 }, { "epoch": 0.4939480242079032, "grad_norm": 1.4191107749938965, "learning_rate": 1.9812932919950892e-05, "loss": 1.0367, "step": 5550 }, { "epoch": 0.4983980064079744, "grad_norm": 1.5782831907272339, "learning_rate": 1.9809553665465705e-05, "loss": 1.2628, "step": 5600 }, { "epoch": 0.5028479886080456, "grad_norm": 1.0732156038284302, "learning_rate": 1.980614445486781e-05, "loss": 0.946, "step": 5650 }, { "epoch": 0.5072979708081168, "grad_norm": 1.3754124641418457, "learning_rate": 1.9802705298568146e-05, "loss": 1.0594, "step": 5700 }, { "epoch": 0.511747953008188, "grad_norm": 1.3374049663543701, "learning_rate": 1.9799236207069107e-05, "loss": 1.16, "step": 5750 }, { "epoch": 0.5161979352082592, "grad_norm": 0.9600222110748291, "learning_rate": 1.9795737190964495e-05, "loss": 1.2729, "step": 5800 }, { "epoch": 0.5206479174083304, "grad_norm": 2.389089345932007, "learning_rate": 1.9792208260939496e-05, "loss": 0.8978, "step": 5850 }, { "epoch": 0.5250978996084016, "grad_norm": 1.7863914966583252, "learning_rate": 1.9788649427770656e-05, "loss": 1.1304, "step": 5900 }, { "epoch": 0.5295478818084728, "grad_norm": 7.10957670211792, "learning_rate": 1.9785060702325817e-05, "loss": 1.4896, "step": 5950 }, { "epoch": 0.533997864008544, "grad_norm": 2.2545292377471924, "learning_rate": 1.978144209556413e-05, "loss": 1.2495, "step": 6000 }, { "epoch": 0.5384478462086152, "grad_norm": 1.3546550273895264, "learning_rate": 1.9777793618535977e-05, "loss": 1.1406, "step": 6050 }, { "epoch": 0.5428978284086864, "grad_norm": 3.378833532333374, "learning_rate": 1.9774115282382972e-05, "loss": 1.2241, "step": 6100 }, { "epoch": 0.5473478106087576, "grad_norm": 17.440778732299805, "learning_rate": 1.9770407098337895e-05, "loss": 1.0015, "step": 6150 }, { "epoch": 0.5517977928088288, "grad_norm": 0.8993490934371948, "learning_rate": 1.9766669077724695e-05, "loss": 1.028, "step": 6200 }, { "epoch": 0.5562477750089, "grad_norm": 9.617663383483887, "learning_rate": 1.9762901231958414e-05, "loss": 1.0712, "step": 6250 }, { "epoch": 0.5606977572089712, "grad_norm": 3.0811498165130615, "learning_rate": 1.9759103572545192e-05, "loss": 1.0916, "step": 6300 }, { "epoch": 0.5651477394090424, "grad_norm": 12.106643676757812, "learning_rate": 1.97552761110822e-05, "loss": 1.1821, "step": 6350 }, { "epoch": 0.5695977216091136, "grad_norm": 1.0771924257278442, "learning_rate": 1.975141885925762e-05, "loss": 1.1795, "step": 6400 }, { "epoch": 0.5740477038091848, "grad_norm": 10.235311508178711, "learning_rate": 1.974753182885061e-05, "loss": 0.9752, "step": 6450 }, { "epoch": 0.578497686009256, "grad_norm": 0.8106779456138611, "learning_rate": 1.9743615031731267e-05, "loss": 0.846, "step": 6500 }, { "epoch": 0.5829476682093272, "grad_norm": 0.7819272875785828, "learning_rate": 1.9739668479860583e-05, "loss": 1.1892, "step": 6550 }, { "epoch": 0.5873976504093984, "grad_norm": 13.416181564331055, "learning_rate": 1.9735692185290415e-05, "loss": 0.9261, "step": 6600 }, { "epoch": 0.5918476326094696, "grad_norm": 1.678322196006775, "learning_rate": 1.9731686160163452e-05, "loss": 1.0477, "step": 6650 }, { "epoch": 0.5962976148095408, "grad_norm": 1.0335452556610107, "learning_rate": 1.972765041671317e-05, "loss": 1.1307, "step": 6700 }, { "epoch": 0.600747597009612, "grad_norm": 11.833608627319336, "learning_rate": 1.972358496726379e-05, "loss": 1.278, "step": 6750 }, { "epoch": 0.6051975792096832, "grad_norm": 1.0137460231781006, "learning_rate": 1.9719489824230268e-05, "loss": 0.9707, "step": 6800 }, { "epoch": 0.6096475614097544, "grad_norm": 1.0877763032913208, "learning_rate": 1.971536500011822e-05, "loss": 1.0194, "step": 6850 }, { "epoch": 0.6140975436098256, "grad_norm": 0.6494864821434021, "learning_rate": 1.9711210507523906e-05, "loss": 1.223, "step": 6900 }, { "epoch": 0.6185475258098968, "grad_norm": 1.7761880159378052, "learning_rate": 1.970702635913419e-05, "loss": 1.1327, "step": 6950 }, { "epoch": 0.622997508009968, "grad_norm": 2.3199312686920166, "learning_rate": 1.9702812567726493e-05, "loss": 1.1433, "step": 7000 }, { "epoch": 0.6274474902100392, "grad_norm": 231.38368225097656, "learning_rate": 1.9698569146168762e-05, "loss": 1.1537, "step": 7050 }, { "epoch": 0.6318974724101104, "grad_norm": 1.9424742460250854, "learning_rate": 1.969429610741943e-05, "loss": 0.8475, "step": 7100 }, { "epoch": 0.6363474546101816, "grad_norm": 1.5187716484069824, "learning_rate": 1.9689993464527363e-05, "loss": 1.1439, "step": 7150 }, { "epoch": 0.6407974368102528, "grad_norm": 0.7318978309631348, "learning_rate": 1.968566123063185e-05, "loss": 0.9806, "step": 7200 }, { "epoch": 0.645247419010324, "grad_norm": 1.2728520631790161, "learning_rate": 1.9681299418962524e-05, "loss": 1.2548, "step": 7250 }, { "epoch": 0.6496974012103952, "grad_norm": 2.5126068592071533, "learning_rate": 1.9676908042839355e-05, "loss": 1.0031, "step": 7300 }, { "epoch": 0.6541473834104664, "grad_norm": 0.7228437662124634, "learning_rate": 1.9672487115672594e-05, "loss": 1.1321, "step": 7350 }, { "epoch": 0.6585973656105376, "grad_norm": 1.3764936923980713, "learning_rate": 1.966803665096273e-05, "loss": 0.9916, "step": 7400 }, { "epoch": 0.6630473478106088, "grad_norm": 1.3645603656768799, "learning_rate": 1.9663556662300454e-05, "loss": 1.0035, "step": 7450 }, { "epoch": 0.66749733001068, "grad_norm": 0.848249077796936, "learning_rate": 1.9659047163366622e-05, "loss": 1.1513, "step": 7500 }, { "epoch": 0.6719473122107511, "grad_norm": 1.5419740676879883, "learning_rate": 1.9654508167932195e-05, "loss": 1.163, "step": 7550 }, { "epoch": 0.6763972944108223, "grad_norm": 22.703840255737305, "learning_rate": 1.9649939689858225e-05, "loss": 1.3058, "step": 7600 }, { "epoch": 0.6808472766108935, "grad_norm": 0.5974789261817932, "learning_rate": 1.9645341743095786e-05, "loss": 0.9678, "step": 7650 }, { "epoch": 0.6852972588109647, "grad_norm": 1.0212652683258057, "learning_rate": 1.9640714341685947e-05, "loss": 1.1219, "step": 7700 }, { "epoch": 0.6897472410110359, "grad_norm": 46.52585983276367, "learning_rate": 1.9636057499759724e-05, "loss": 1.0244, "step": 7750 }, { "epoch": 0.6941972232111071, "grad_norm": 1.4065423011779785, "learning_rate": 1.9631371231538035e-05, "loss": 1.0998, "step": 7800 }, { "epoch": 0.6986472054111783, "grad_norm": 0.6876867413520813, "learning_rate": 1.9626655551331665e-05, "loss": 1.2191, "step": 7850 }, { "epoch": 0.7030971876112495, "grad_norm": 1.2235878705978394, "learning_rate": 1.962191047354121e-05, "loss": 1.1293, "step": 7900 }, { "epoch": 0.7075471698113207, "grad_norm": 1.5235185623168945, "learning_rate": 1.9617136012657045e-05, "loss": 1.0222, "step": 7950 }, { "epoch": 0.7119971520113919, "grad_norm": 2.0007951259613037, "learning_rate": 1.9612332183259267e-05, "loss": 1.0209, "step": 8000 }, { "epoch": 0.7164471342114631, "grad_norm": 0.7732981443405151, "learning_rate": 1.9607499000017668e-05, "loss": 1.1007, "step": 8050 }, { "epoch": 0.7208971164115343, "grad_norm": 3.4698424339294434, "learning_rate": 1.9602636477691674e-05, "loss": 1.253, "step": 8100 }, { "epoch": 0.7253470986116055, "grad_norm": 0.8018836379051208, "learning_rate": 1.9597744631130298e-05, "loss": 1.186, "step": 8150 }, { "epoch": 0.7297970808116767, "grad_norm": 1.1866015195846558, "learning_rate": 1.9592823475272117e-05, "loss": 1.1985, "step": 8200 }, { "epoch": 0.7342470630117479, "grad_norm": 1.5124443769454956, "learning_rate": 1.9587873025145202e-05, "loss": 0.9594, "step": 8250 }, { "epoch": 0.7386970452118191, "grad_norm": 10.030634880065918, "learning_rate": 1.958289329586709e-05, "loss": 1.4065, "step": 8300 }, { "epoch": 0.7431470274118903, "grad_norm": 1.049864411354065, "learning_rate": 1.957788430264472e-05, "loss": 1.0799, "step": 8350 }, { "epoch": 0.7475970096119615, "grad_norm": 0.6957271695137024, "learning_rate": 1.9572846060774403e-05, "loss": 1.1008, "step": 8400 }, { "epoch": 0.7520469918120327, "grad_norm": 1.6415716409683228, "learning_rate": 1.9567880221530444e-05, "loss": 1.1157, "step": 8450 }, { "epoch": 0.7564969740121039, "grad_norm": 5.416161060333252, "learning_rate": 1.9562784112813907e-05, "loss": 1.156, "step": 8500 }, { "epoch": 0.7609469562121751, "grad_norm": 22.172027587890625, "learning_rate": 1.955765880156191e-05, "loss": 1.3024, "step": 8550 }, { "epoch": 0.7653969384122463, "grad_norm": 1.2330561876296997, "learning_rate": 1.9552504303425986e-05, "loss": 1.1647, "step": 8600 }, { "epoch": 0.7698469206123175, "grad_norm": 1.1770780086517334, "learning_rate": 1.9547320634146767e-05, "loss": 0.9177, "step": 8650 }, { "epoch": 0.7742969028123887, "grad_norm": 3.6036808490753174, "learning_rate": 1.9542107809553995e-05, "loss": 1.147, "step": 8700 }, { "epoch": 0.7787468850124599, "grad_norm": 0.9867936968803406, "learning_rate": 1.9536865845566425e-05, "loss": 0.9828, "step": 8750 }, { "epoch": 0.7831968672125311, "grad_norm": 17.20427894592285, "learning_rate": 1.953159475819181e-05, "loss": 1.127, "step": 8800 }, { "epoch": 0.7876468494126023, "grad_norm": 0.6954526305198669, "learning_rate": 1.952629456352683e-05, "loss": 0.8992, "step": 8850 }, { "epoch": 0.7920968316126735, "grad_norm": 0.7507011890411377, "learning_rate": 1.9520965277757054e-05, "loss": 1.2564, "step": 8900 }, { "epoch": 0.7965468138127447, "grad_norm": 2.2453718185424805, "learning_rate": 1.9515606917156904e-05, "loss": 0.984, "step": 8950 }, { "epoch": 0.8009967960128159, "grad_norm": 1.342645287513733, "learning_rate": 1.9510219498089563e-05, "loss": 1.1199, "step": 9000 }, { "epoch": 0.8054467782128871, "grad_norm": 1.5172566175460815, "learning_rate": 1.950480303700697e-05, "loss": 0.9538, "step": 9050 }, { "epoch": 0.8098967604129583, "grad_norm": 1.8109244108200073, "learning_rate": 1.9499357550449745e-05, "loss": 1.0505, "step": 9100 }, { "epoch": 0.8143467426130295, "grad_norm": 14.07536792755127, "learning_rate": 1.949388305504715e-05, "loss": 1.1659, "step": 9150 }, { "epoch": 0.8187967248131007, "grad_norm": 0.8219563364982605, "learning_rate": 1.9488379567517032e-05, "loss": 1.1572, "step": 9200 }, { "epoch": 0.8232467070131719, "grad_norm": 0.7456687688827515, "learning_rate": 1.9482847104665765e-05, "loss": 1.0114, "step": 9250 }, { "epoch": 0.8276966892132431, "grad_norm": 2.4477434158325195, "learning_rate": 1.9477285683388218e-05, "loss": 1.1338, "step": 9300 }, { "epoch": 0.8321466714133143, "grad_norm": 1.7693390846252441, "learning_rate": 1.947169532066769e-05, "loss": 0.9649, "step": 9350 }, { "epoch": 0.8365966536133855, "grad_norm": 13.62708568572998, "learning_rate": 1.946607603357585e-05, "loss": 1.0521, "step": 9400 }, { "epoch": 0.8410466358134567, "grad_norm": 2.1173110008239746, "learning_rate": 1.9460427839272714e-05, "loss": 1.4169, "step": 9450 }, { "epoch": 0.8454966180135279, "grad_norm": 0.8660404682159424, "learning_rate": 1.9454750755006557e-05, "loss": 0.8661, "step": 9500 }, { "epoch": 0.8499466002135991, "grad_norm": 9.612180709838867, "learning_rate": 1.9449044798113888e-05, "loss": 1.1864, "step": 9550 }, { "epoch": 0.8543965824136703, "grad_norm": 0.7924606204032898, "learning_rate": 1.9443309986019377e-05, "loss": 1.1295, "step": 9600 }, { "epoch": 0.8588465646137415, "grad_norm": 0.5419692397117615, "learning_rate": 1.9437546336235824e-05, "loss": 0.9944, "step": 9650 }, { "epoch": 0.8632965468138127, "grad_norm": 2.778602123260498, "learning_rate": 1.943175386636408e-05, "loss": 1.3265, "step": 9700 }, { "epoch": 0.8677465290138839, "grad_norm": 2.985241413116455, "learning_rate": 1.942593259409301e-05, "loss": 1.0673, "step": 9750 }, { "epoch": 0.8721965112139551, "grad_norm": 3.0507829189300537, "learning_rate": 1.9420082537199437e-05, "loss": 1.1267, "step": 9800 }, { "epoch": 0.8766464934140263, "grad_norm": 1.0891329050064087, "learning_rate": 1.9414203713548086e-05, "loss": 1.1279, "step": 9850 }, { "epoch": 0.8810964756140975, "grad_norm": 2.639254570007324, "learning_rate": 1.9408296141091533e-05, "loss": 1.3157, "step": 9900 }, { "epoch": 0.8855464578141687, "grad_norm": 2.320467710494995, "learning_rate": 1.9402359837870128e-05, "loss": 0.9687, "step": 9950 }, { "epoch": 0.8899964400142399, "grad_norm": 1.3457088470458984, "learning_rate": 1.939639482201198e-05, "loss": 0.8925, "step": 10000 }, { "epoch": 0.8944464222143111, "grad_norm": 1.1230852603912354, "learning_rate": 1.9390401111732868e-05, "loss": 1.0535, "step": 10050 }, { "epoch": 0.8988964044143823, "grad_norm": 2.784928798675537, "learning_rate": 1.93843787253362e-05, "loss": 1.1733, "step": 10100 }, { "epoch": 0.9033463866144535, "grad_norm": 1.3313930034637451, "learning_rate": 1.9378327681212954e-05, "loss": 1.0189, "step": 10150 }, { "epoch": 0.9077963688145247, "grad_norm": 1.4785683155059814, "learning_rate": 1.9372247997841623e-05, "loss": 1.1516, "step": 10200 }, { "epoch": 0.912246351014596, "grad_norm": 4.558147430419922, "learning_rate": 1.936613969378815e-05, "loss": 1.3025, "step": 10250 }, { "epoch": 0.9166963332146671, "grad_norm": 1.1733040809631348, "learning_rate": 1.93600027877059e-05, "loss": 1.1425, "step": 10300 }, { "epoch": 0.9211463154147383, "grad_norm": 11.623977661132812, "learning_rate": 1.935383729833555e-05, "loss": 1.0945, "step": 10350 }, { "epoch": 0.9255962976148095, "grad_norm": 2.0317745208740234, "learning_rate": 1.9347643244505098e-05, "loss": 0.9415, "step": 10400 }, { "epoch": 0.9300462798148807, "grad_norm": 1.0777006149291992, "learning_rate": 1.934142064512974e-05, "loss": 0.9, "step": 10450 }, { "epoch": 0.934496262014952, "grad_norm": 0.9465547800064087, "learning_rate": 1.9335169519211873e-05, "loss": 1.1661, "step": 10500 }, { "epoch": 0.9389462442150232, "grad_norm": 3.3057515621185303, "learning_rate": 1.9328889885840982e-05, "loss": 1.0773, "step": 10550 }, { "epoch": 0.9433962264150944, "grad_norm": 1.0848942995071411, "learning_rate": 1.932258176419362e-05, "loss": 0.9374, "step": 10600 }, { "epoch": 0.9478462086151656, "grad_norm": 0.6923954486846924, "learning_rate": 1.931624517353334e-05, "loss": 0.8315, "step": 10650 }, { "epoch": 0.9522961908152368, "grad_norm": 0.835031270980835, "learning_rate": 1.930988013321062e-05, "loss": 1.1184, "step": 10700 }, { "epoch": 0.956746173015308, "grad_norm": 0.8372194170951843, "learning_rate": 1.9303486662662824e-05, "loss": 0.8965, "step": 10750 }, { "epoch": 0.9611961552153792, "grad_norm": 1.5202910900115967, "learning_rate": 1.9297064781414142e-05, "loss": 1.1413, "step": 10800 }, { "epoch": 0.9656461374154504, "grad_norm": 17.712987899780273, "learning_rate": 1.929061450907551e-05, "loss": 0.9889, "step": 10850 }, { "epoch": 0.9700961196155216, "grad_norm": 0.6960763931274414, "learning_rate": 1.9284135865344576e-05, "loss": 1.2101, "step": 10900 }, { "epoch": 0.9745461018155928, "grad_norm": 0.6224497556686401, "learning_rate": 1.927762887000562e-05, "loss": 0.853, "step": 10950 }, { "epoch": 0.978996084015664, "grad_norm": 2.1982674598693848, "learning_rate": 1.9271093542929508e-05, "loss": 1.2139, "step": 11000 }, { "epoch": 0.9834460662157352, "grad_norm": 3.0297718048095703, "learning_rate": 1.9264529904073614e-05, "loss": 1.201, "step": 11050 }, { "epoch": 0.9878960484158064, "grad_norm": 1.381274938583374, "learning_rate": 1.9257937973481778e-05, "loss": 1.1065, "step": 11100 }, { "epoch": 0.9923460306158776, "grad_norm": 20.468204498291016, "learning_rate": 1.925131777128424e-05, "loss": 1.3885, "step": 11150 }, { "epoch": 0.9967960128159488, "grad_norm": 0.8038458824157715, "learning_rate": 1.9244669317697563e-05, "loss": 1.4847, "step": 11200 }, { "epoch": 1.0, "eval_f1_homophobe": 0.7517241379310344, "eval_f1_macro": 0.5682024245488293, "eval_f1_micro": 0.6462405673243022, "eval_f1_otherhate": 0.5995331777259086, "eval_f1_racist": 0.6786310517529215, "eval_f1_religion": 0.20270270270270271, "eval_f1_sexist": 0.608421052631579, "eval_loss": 1.4235976934432983, "eval_roc_homophobe": 0.8922838430419266, "eval_roc_macro": 0.7886598443962021, "eval_roc_otherhate": 0.8324156169761214, "eval_roc_racist": 0.7609558549233864, "eval_roc_religion": 0.635975612667281, "eval_roc_sexist": 0.8216682943722944, "eval_runtime": 20.8847, "eval_samples_per_second": 239.41, "eval_steps_per_second": 7.517, "step": 11236 } ], "logging_steps": 50, "max_steps": 89888, "num_input_tokens_seen": 0, "num_train_epochs": 8, "save_steps": 500, "stateful_callbacks": { "EarlyStoppingCallback": { "args": { "early_stopping_patience": 3, "early_stopping_threshold": 0.0 }, "attributes": { "early_stopping_patience_counter": 0 } }, "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": false }, "attributes": {} } }, "total_flos": 0.0, "train_batch_size": 12, "trial_name": null, "trial_params": null }