| { | |
| "best_metric": 0.06251987814903259, | |
| "best_model_checkpoint": "saves/psy-course-doc/Llama-3.1-8B-Instruct/train/fold9/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": 0.16012704372406006, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.2188, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 0.1626744419336319, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.2227, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 0.18549920618534088, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.1686, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 0.17375493049621582, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.2938, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 0.17902612686157227, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.2091, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 0.18505369126796722, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.1754, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 0.19533436000347137, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.1805, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 0.18160583078861237, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.2426, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.16058214008808136, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.1368, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 0.14139407873153687, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.2259, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.14175167679786682, | |
| "eval_runtime": 22.9155, | |
| "eval_samples_per_second": 1.964, | |
| "eval_steps_per_second": 1.964, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 0.15317194163799286, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.1597, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 0.17080363631248474, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.1835, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 0.16679717600345612, | |
| "learning_rate": 0.0001, | |
| "loss": 0.139, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.16384510695934296, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.1513, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.1557767689228058, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.1007, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.14455045759677887, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.1663, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.1448976993560791, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.1246, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.12634430825710297, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.0858, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.11505317687988281, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.1494, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.10868946462869644, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.172, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.08925747126340866, | |
| "eval_runtime": 22.8471, | |
| "eval_samples_per_second": 1.97, | |
| "eval_steps_per_second": 1.97, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.09257800132036209, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.0908, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.09200786799192429, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.0843, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.09939802438020706, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.1106, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.10130990296602249, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.1202, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.09145660698413849, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.1106, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 0.20054078102111816, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.1968, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.10766153782606125, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.102, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.10508635640144348, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.1394, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.10992046445608139, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.1035, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.09355330467224121, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.0791, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.07540746033191681, | |
| "eval_runtime": 22.8428, | |
| "eval_samples_per_second": 1.97, | |
| "eval_steps_per_second": 1.97, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.11442509293556213, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.1317, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.11588981002569199, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.124, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.10404524207115173, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.1212, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.09653621166944504, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.1008, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.07260455191135406, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.0485, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.09970170259475708, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.1211, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.07373712211847305, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.0547, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.0925954282283783, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.1459, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.09566950052976608, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.122, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.08995156735181808, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.0414, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.06949092447757721, | |
| "eval_runtime": 22.9091, | |
| "eval_samples_per_second": 1.964, | |
| "eval_steps_per_second": 1.964, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.07674718648195267, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0266, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.08287996053695679, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.0589, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.0897073820233345, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.0993, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.10945425927639008, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.1523, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.07625408470630646, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.0327, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.08506359905004501, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.0575, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.08949103206396103, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.1164, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.09872898459434509, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.1056, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.07700581848621368, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.0655, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.10579069703817368, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.0704, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.06626862287521362, | |
| "eval_runtime": 22.8666, | |
| "eval_samples_per_second": 1.968, | |
| "eval_steps_per_second": 1.968, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 0.16567502915859222, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.1483, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.07487107068300247, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.0686, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.0772494301199913, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.097, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.08777953684329987, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.0962, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.0702255591750145, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0511, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.09427014738321304, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.1301, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.09751700609922409, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.0769, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.09331370890140533, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.0894, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.10984798520803452, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.0949, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.08716970682144165, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.0784, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.0646897703409195, | |
| "eval_runtime": 22.843, | |
| "eval_samples_per_second": 1.97, | |
| "eval_steps_per_second": 1.97, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.08923397958278656, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.0378, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.09308641403913498, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.0806, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.07544505596160889, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.0507, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.11075833439826965, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.0507, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.1257985383272171, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.0972, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.0933673232793808, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.0677, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.09619047492742538, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.1156, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.11217456310987473, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.1447, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.11414963006973267, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0922, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.10665580630302429, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.1002, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.06409724801778793, | |
| "eval_runtime": 22.8363, | |
| "eval_samples_per_second": 1.971, | |
| "eval_steps_per_second": 1.971, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.09320751577615738, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.0748, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.09927953779697418, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.1017, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.1172712966799736, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.1052, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.0946318581700325, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.0497, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.09094702452421188, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.0543, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 0.16316764056682587, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.1286, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.07997765392065048, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0485, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.1100013479590416, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.1499, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.08684566617012024, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.0454, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.10784677416086197, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.1164, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.06294846534729004, | |
| "eval_runtime": 22.8349, | |
| "eval_samples_per_second": 1.971, | |
| "eval_steps_per_second": 1.971, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.08588241040706635, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.0732, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.0709327757358551, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0297, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.09673380851745605, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.1105, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.09051958471536636, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.0403, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.09911269694566727, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0781, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.07569113373756409, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.0444, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.11394467204809189, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.0689, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.11975882947444916, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.156, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.1024811640381813, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.1163, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.10779350996017456, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0822, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.06290505826473236, | |
| "eval_runtime": 22.8598, | |
| "eval_samples_per_second": 1.969, | |
| "eval_steps_per_second": 1.969, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.13391198217868805, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.1135, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.09439925849437714, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.0635, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.07956387847661972, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.0431, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.10458944737911224, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.0868, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.0939633771777153, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.0522, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.09950806945562363, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.054, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.11619756370782852, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.0962, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.10245565325021744, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.0676, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.10659981518983841, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.0655, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.09897475689649582, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.0583, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.06251987814903259, | |
| "eval_runtime": 22.8549, | |
| "eval_samples_per_second": 1.969, | |
| "eval_steps_per_second": 1.969, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 0.07174082100391388, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.025, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 0.14929655194282532, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.0668, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.10811996459960938, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.0871, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.08908949047327042, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0341, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.1073828712105751, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0994, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.0953696072101593, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0749, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.09153662621974945, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.0479, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.0921344980597496, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.0713, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.10496306419372559, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.0785, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.07941649854183197, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0488, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.06252405047416687, | |
| "eval_runtime": 22.8465, | |
| "eval_samples_per_second": 1.97, | |
| "eval_steps_per_second": 1.97, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.09306307137012482, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0711, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.10954549908638, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0794, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.10445868223905563, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.0809, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.09497830271720886, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.052, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.09065257012844086, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0602, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.09563349932432175, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.0427, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.08353404700756073, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.0449, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.10227520763874054, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0745, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.10729608684778214, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.0839, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.10004584491252899, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.0822, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.0625731498003006, | |
| "eval_runtime": 22.8373, | |
| "eval_samples_per_second": 1.97, | |
| "eval_steps_per_second": 1.97, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.11356940865516663, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.12, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.09182242304086685, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0783, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.09101668745279312, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.08, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.09921537339687347, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0667, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.09743266552686691, | |
| "learning_rate": 0.0, | |
| "loss": 0.0659, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.4942709034234675e+17, | |
| "train_loss": 0.09839795534312724, | |
| "train_runtime": 3683.0658, | |
| "train_samples_per_second": 0.55, | |
| "train_steps_per_second": 0.034 | |
| } | |
| ], | |
| "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": 1.4942709034234675e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |