| { | |
| "best_metric": 0.016363559290766716, | |
| "best_model_checkpoint": "saves/psy-course-doc/Llama-3.1-8B-Instruct/train/fold3/checkpoint-100", | |
| "epoch": 4.938271604938271, | |
| "eval_steps": 10, | |
| "global_step": 125, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03950617283950617, | |
| "grad_norm": 0.11982958763837814, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.1009, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 0.18020769953727722, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.1311, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 0.17260421812534332, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.1305, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 0.1354549378156662, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.1094, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 0.10506410896778107, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.0996, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 0.14751164615154266, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.11, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 0.135710671544075, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.1059, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 0.13157124817371368, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.097, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.13530097901821136, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.1037, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 0.12781627476215363, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.0899, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.08468911796808243, | |
| "eval_runtime": 29.6804, | |
| "eval_samples_per_second": 1.516, | |
| "eval_steps_per_second": 1.516, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 0.11392076313495636, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.0682, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 0.1307547390460968, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.0791, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 0.1341903805732727, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0647, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.12305284291505814, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.0521, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.14141477644443512, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.0561, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.1190507784485817, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.0482, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.12106993794441223, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.0462, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.13914185762405396, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.0434, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.0968879908323288, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.0377, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.07901491969823837, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.0334, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.03550420328974724, | |
| "eval_runtime": 29.5606, | |
| "eval_samples_per_second": 1.522, | |
| "eval_steps_per_second": 1.522, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.07772660255432129, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.0321, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.08289165794849396, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.0341, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.06809127330780029, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.0245, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.08088243007659912, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.0298, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.09069989621639252, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.0291, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 0.14729762077331543, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.0465, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.06314294785261154, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0233, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.07074397057294846, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.0269, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.0783693790435791, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.0264, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.07551830261945724, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.0204, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.024371623992919922, | |
| "eval_runtime": 29.5682, | |
| "eval_samples_per_second": 1.522, | |
| "eval_steps_per_second": 1.522, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.07203515619039536, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.0225, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.082725889980793, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.0252, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.060164906084537506, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.0188, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.07446308434009552, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.0236, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.06977657973766327, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.0225, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.06203348562121391, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.0181, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.07342709600925446, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.0221, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.07633589953184128, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.0177, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.056108083575963974, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.0175, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.06389579176902771, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.0215, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.020678821951150894, | |
| "eval_runtime": 29.5921, | |
| "eval_samples_per_second": 1.521, | |
| "eval_steps_per_second": 1.521, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.0723925456404686, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.015, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.05432877689599991, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.0158, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.05594627931714058, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.0164, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.05370336398482323, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.0226, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.07103245705366135, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.0176, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.06544952839612961, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.0186, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.05212434381246567, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.0158, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.06580112874507904, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.0181, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.060175977647304535, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.0169, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.05927619710564613, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.0173, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.018583130091428757, | |
| "eval_runtime": 29.5769, | |
| "eval_samples_per_second": 1.521, | |
| "eval_steps_per_second": 1.521, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 0.11036009341478348, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.0289, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.05153071880340576, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.016, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.06391610205173492, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.0139, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.07181736081838608, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.014, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.04309231787919998, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0123, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.053510796278715134, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.0143, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.06065421551465988, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.019, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.06207943707704544, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.0158, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.06607165187597275, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.0185, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.06022383272647858, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.014, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.01788869872689247, | |
| "eval_runtime": 29.5983, | |
| "eval_samples_per_second": 1.52, | |
| "eval_steps_per_second": 1.52, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.0726022720336914, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.017, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.058323588222265244, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.0119, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.06410640478134155, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.0147, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.053326427936553955, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.0137, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.06104976311326027, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.0139, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.06476392596960068, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.0167, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.06204933673143387, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.0137, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.058300163596868515, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.0137, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.058873895555734634, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0169, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.07083891332149506, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.0131, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.01707731932401657, | |
| "eval_runtime": 29.5547, | |
| "eval_samples_per_second": 1.523, | |
| "eval_steps_per_second": 1.523, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.06999720633029938, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.0141, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.06281733512878418, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.0129, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.061437204480171204, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.0106, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.05753050744533539, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.0183, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.07456520944833755, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.0162, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 0.09577728062868118, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.0246, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.05529274418950081, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0119, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.039042335003614426, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.0137, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.04296581447124481, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.0109, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.06321784853935242, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.0152, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.016468888148665428, | |
| "eval_runtime": 29.5492, | |
| "eval_samples_per_second": 1.523, | |
| "eval_steps_per_second": 1.523, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.05038125813007355, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.0128, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.046389661729335785, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0128, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.06602297723293304, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0116, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.047070205211639404, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.0111, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.04912222921848297, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0106, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.04519559070467949, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.0141, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.05045366659760475, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.0096, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.070041723549366, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.0125, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.05587673932313919, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.0107, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.05657823756337166, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.014, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.017023487016558647, | |
| "eval_runtime": 29.5202, | |
| "eval_samples_per_second": 1.524, | |
| "eval_steps_per_second": 1.524, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.0667024478316307, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.0143, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.04744872823357582, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.0118, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.07030303031206131, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.0121, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.05097898840904236, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.0141, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.06667334586381912, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.0151, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.06494573503732681, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0108, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.0436541885137558, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0108, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.06152384355664253, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.0125, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.05768159404397011, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.0126, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.05780719220638275, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.0106, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.016363559290766716, | |
| "eval_runtime": 29.4826, | |
| "eval_samples_per_second": 1.526, | |
| "eval_steps_per_second": 1.526, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 0.061449840664863586, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.0151, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 0.08285302668809891, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.019, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.039812907576560974, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.0105, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.04446830600500107, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0095, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.04810008034110069, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0139, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.04866350069642067, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0097, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.04433601349592209, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.0102, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.04722043126821518, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.015, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.0726417750120163, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.015, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.0651179701089859, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0133, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.01652962900698185, | |
| "eval_runtime": 29.4288, | |
| "eval_samples_per_second": 1.529, | |
| "eval_steps_per_second": 1.529, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.046997252851724625, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0082, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.05112281069159508, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0084, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.05031479522585869, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.0122, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.05058969557285309, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.0142, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.047341667115688324, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0112, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.03792574256658554, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.01, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.04679827392101288, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.0119, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.042794372886419296, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0095, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.05135497450828552, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.0134, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.04792887717485428, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.013, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.01643712818622589, | |
| "eval_runtime": 29.3583, | |
| "eval_samples_per_second": 1.533, | |
| "eval_steps_per_second": 1.533, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.054526232182979584, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.01, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.043358638882637024, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0108, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.05204227939248085, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.0111, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.040465980768203735, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0091, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.048052042722702026, | |
| "learning_rate": 0.0, | |
| "loss": 0.0129, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.9234926316191744e+17, | |
| "train_loss": 0.026285644583404065, | |
| "train_runtime": 5836.1157, | |
| "train_samples_per_second": 0.347, | |
| "train_steps_per_second": 0.021 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 125, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.9234926316191744e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |