| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 2.0647521018981934, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1825, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 1.9963816404342651, | |
| "learning_rate": 8e-05, | |
| "loss": 1.1448, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 1.82085120677948, | |
| "learning_rate": 0.00012, | |
| "loss": 1.1246, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.3797991275787354, | |
| "learning_rate": 0.00016, | |
| "loss": 0.9428, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.0062527656555176, | |
| "learning_rate": 0.0002, | |
| "loss": 0.8782, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.6854935884475708, | |
| "learning_rate": 0.00019994532573409262, | |
| "loss": 0.8182, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 0.6088404059410095, | |
| "learning_rate": 0.00019978136272187747, | |
| "loss": 0.7337, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.5058557391166687, | |
| "learning_rate": 0.00019950829025450114, | |
| "loss": 0.7606, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.4383871853351593, | |
| "learning_rate": 0.00019912640693269752, | |
| "loss": 0.7934, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.3763003349304199, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 0.7623, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.3510531187057495, | |
| "learning_rate": 0.00019803799658748094, | |
| "loss": 0.7043, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.3534976541996002, | |
| "learning_rate": 0.0001973326597248006, | |
| "loss": 0.7702, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.33237403631210327, | |
| "learning_rate": 0.00019652089102773488, | |
| "loss": 0.7074, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.3145230710506439, | |
| "learning_rate": 0.00019560357815343577, | |
| "loss": 0.6778, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.32996928691864014, | |
| "learning_rate": 0.00019458172417006347, | |
| "loss": 0.7032, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.32944002747535706, | |
| "learning_rate": 0.0001934564464599461, | |
| "loss": 0.7088, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.3067278265953064, | |
| "learning_rate": 0.00019222897549773848, | |
| "loss": 0.6803, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.3252066671848297, | |
| "learning_rate": 0.00019090065350491626, | |
| "loss": 0.6947, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.31279104948043823, | |
| "learning_rate": 0.00018947293298207635, | |
| "loss": 0.6402, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.30892765522003174, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 0.6519, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.29156580567359924, | |
| "learning_rate": 0.00018632564809575742, | |
| "loss": 0.6747, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.2854658365249634, | |
| "learning_rate": 0.00018460952524209355, | |
| "loss": 0.6556, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 0.2657647430896759, | |
| "learning_rate": 0.00018280088311480201, | |
| "loss": 0.6128, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.29050764441490173, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.6674, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.29195520281791687, | |
| "learning_rate": 0.00017891405093963938, | |
| "loss": 0.63, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 0.30334123969078064, | |
| "learning_rate": 0.00017684011108568592, | |
| "loss": 0.5533, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.30424660444259644, | |
| "learning_rate": 0.0001746821476984154, | |
| "loss": 0.5549, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 0.28369948267936707, | |
| "learning_rate": 0.00017244252047910892, | |
| "loss": 0.5415, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.2778087258338928, | |
| "learning_rate": 0.00017012367842724887, | |
| "loss": 0.5382, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.2843495011329651, | |
| "learning_rate": 0.00016772815716257412, | |
| "loss": 0.5388, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.28178870677948, | |
| "learning_rate": 0.00016525857615241687, | |
| "loss": 0.5126, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.28643131256103516, | |
| "learning_rate": 0.0001627176358473537, | |
| "loss": 0.5167, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.28843289613723755, | |
| "learning_rate": 0.00016010811472830252, | |
| "loss": 0.5099, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 0.28687819838523865, | |
| "learning_rate": 0.00015743286626829437, | |
| "loss": 0.5185, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.28858163952827454, | |
| "learning_rate": 0.00015469481581224272, | |
| "loss": 0.5279, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 0.2918362021446228, | |
| "learning_rate": 0.00015189695737812152, | |
| "loss": 0.4788, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.291130930185318, | |
| "learning_rate": 0.00014904235038305083, | |
| "loss": 0.4856, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.2943766415119171, | |
| "learning_rate": 0.0001461341162978688, | |
| "loss": 0.4615, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.297374427318573, | |
| "learning_rate": 0.00014317543523384928, | |
| "loss": 0.4583, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.30772677063941956, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 0.4859, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.29820892214775085, | |
| "learning_rate": 0.00013711972489182208, | |
| "loss": 0.4768, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6800000000000002, | |
| "grad_norm": 0.29821571707725525, | |
| "learning_rate": 0.00013402931744416433, | |
| "loss": 0.4891, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.2966136634349823, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.4794, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.3031865358352661, | |
| "learning_rate": 0.00012774029087618446, | |
| "loss": 0.4853, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.2918758690357208, | |
| "learning_rate": 0.00012454854871407994, | |
| "loss": 0.4523, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.8399999999999999, | |
| "grad_norm": 0.3020419180393219, | |
| "learning_rate": 0.0001213299630743747, | |
| "loss": 0.4668, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.30048614740371704, | |
| "learning_rate": 0.000118088053433211, | |
| "loss": 0.4468, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.3022134304046631, | |
| "learning_rate": 0.0001148263647711842, | |
| "loss": 0.4636, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.3102932572364807, | |
| "learning_rate": 0.00011154846369695863, | |
| "loss": 0.4762, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.29303210973739624, | |
| "learning_rate": 0.00010825793454723325, | |
| "loss": 0.4109, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.29739847779273987, | |
| "learning_rate": 0.00010495837546732224, | |
| "loss": 0.3787, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.29793456196784973, | |
| "learning_rate": 0.00010165339447663587, | |
| "loss": 0.3708, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 0.31656354665756226, | |
| "learning_rate": 9.834660552336415e-05, | |
| "loss": 0.3855, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.30371928215026855, | |
| "learning_rate": 9.504162453267777e-05, | |
| "loss": 0.3873, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 0.31064578890800476, | |
| "learning_rate": 9.174206545276677e-05, | |
| "loss": 0.3595, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.29833751916885376, | |
| "learning_rate": 8.845153630304139e-05, | |
| "loss": 0.366, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.2800000000000002, | |
| "grad_norm": 0.3072323501110077, | |
| "learning_rate": 8.517363522881579e-05, | |
| "loss": 0.3621, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.29597151279449463, | |
| "learning_rate": 8.191194656678904e-05, | |
| "loss": 0.3519, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 0.3137385845184326, | |
| "learning_rate": 7.867003692562534e-05, | |
| "loss": 0.3519, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.3161297142505646, | |
| "learning_rate": 7.54514512859201e-05, | |
| "loss": 0.3502, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 0.3286161720752716, | |
| "learning_rate": 7.225970912381556e-05, | |
| "loss": 0.3518, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.3324306905269623, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.357, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "grad_norm": 0.3510464131832123, | |
| "learning_rate": 6.59706825558357e-05, | |
| "loss": 0.379, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.3358787000179291, | |
| "learning_rate": 6.28802751081779e-05, | |
| "loss": 0.3399, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 0.330174058675766, | |
| "learning_rate": 5.983045753470308e-05, | |
| "loss": 0.3465, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.31485724449157715, | |
| "learning_rate": 5.6824564766150726e-05, | |
| "loss": 0.3249, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 0.3314495384693146, | |
| "learning_rate": 5.386588370213124e-05, | |
| "loss": 0.3565, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.334344744682312, | |
| "learning_rate": 5.095764961694922e-05, | |
| "loss": 0.344, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 0.330770343542099, | |
| "learning_rate": 4.810304262187852e-05, | |
| "loss": 0.3424, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.32071828842163086, | |
| "learning_rate": 4.530518418775733e-05, | |
| "loss": 0.3396, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 0.32281526923179626, | |
| "learning_rate": 4.256713373170564e-05, | |
| "loss": 0.343, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.3217308223247528, | |
| "learning_rate": 3.9891885271697496e-05, | |
| "loss": 0.3142, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 0.3167681097984314, | |
| "learning_rate": 3.7282364152646297e-05, | |
| "loss": 0.3351, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.3280259072780609, | |
| "learning_rate": 3.4741423847583134e-05, | |
| "loss": 0.316, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.32448863983154297, | |
| "learning_rate": 3.227184283742591e-05, | |
| "loss": 0.3195, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 0.33512407541275024, | |
| "learning_rate": 2.9876321572751144e-05, | |
| "loss": 0.2814, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 0.33603280782699585, | |
| "learning_rate": 2.7557479520891104e-05, | |
| "loss": 0.2981, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 0.33797961473464966, | |
| "learning_rate": 2.5317852301584643e-05, | |
| "loss": 0.293, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 0.3338729739189148, | |
| "learning_rate": 2.315988891431412e-05, | |
| "loss": 0.2988, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.3124498128890991, | |
| "learning_rate": 2.1085949060360654e-05, | |
| "loss": 0.2956, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "grad_norm": 0.30723869800567627, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.2762, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2800000000000002, | |
| "grad_norm": 0.30877774953842163, | |
| "learning_rate": 1.7199116885197995e-05, | |
| "loss": 0.2811, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "grad_norm": 0.3040075898170471, | |
| "learning_rate": 1.5390474757906446e-05, | |
| "loss": 0.2843, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 0.29099854826927185, | |
| "learning_rate": 1.3674351904242611e-05, | |
| "loss": 0.2823, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 0.29945293068885803, | |
| "learning_rate": 1.2052624879351104e-05, | |
| "loss": 0.3007, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 0.29691964387893677, | |
| "learning_rate": 1.0527067017923654e-05, | |
| "loss": 0.2866, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 0.29864245653152466, | |
| "learning_rate": 9.09934649508375e-06, | |
| "loss": 0.2946, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "grad_norm": 0.29509297013282776, | |
| "learning_rate": 7.771024502261526e-06, | |
| "loss": 0.2717, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "grad_norm": 0.3103945851325989, | |
| "learning_rate": 6.543553540053926e-06, | |
| "loss": 0.287, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 0.2974102795124054, | |
| "learning_rate": 5.418275829936537e-06, | |
| "loss": 0.2815, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "grad_norm": 0.3041694760322571, | |
| "learning_rate": 4.3964218465642355e-06, | |
| "loss": 0.2905, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 0.2940130829811096, | |
| "learning_rate": 3.4791089722651436e-06, | |
| "loss": 0.273, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.7199999999999998, | |
| "grad_norm": 0.2969824969768524, | |
| "learning_rate": 2.667340275199426e-06, | |
| "loss": 0.2691, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.30998706817626953, | |
| "learning_rate": 1.9620034125190644e-06, | |
| "loss": 0.2984, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 0.31238672137260437, | |
| "learning_rate": 1.3638696597277679e-06, | |
| "loss": 0.2856, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 0.29781845211982727, | |
| "learning_rate": 8.735930673024806e-07, | |
| "loss": 0.2911, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "grad_norm": 0.29917433857917786, | |
| "learning_rate": 4.917097454988584e-07, | |
| "loss": 0.278, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 0.30273205041885376, | |
| "learning_rate": 2.1863727812254653e-07, | |
| "loss": 0.2811, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 0.3171875774860382, | |
| "learning_rate": 5.467426590739511e-08, | |
| "loss": 0.2876, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.3021162450313568, | |
| "learning_rate": 0.0, | |
| "loss": 0.2926, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.5000238750511923e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |