| { | |
| "best_metric": 0.08211181312799454, | |
| "best_model_checkpoint": "saves/psy-course-info-chain/Llama3-OpenBioLLM-8B/train/fold8/checkpoint-80", | |
| "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": 1.102138876914978, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.3878, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 1.0834274291992188, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.2834, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 1.1702054738998413, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.3205, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 1.207353949546814, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.3197, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 1.0222070217132568, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.3264, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 1.3054327964782715, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.3549, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 1.307969093322754, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.2904, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 1.0337932109832764, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.227, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 0.9669976830482483, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.2844, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 0.9556586742401123, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.1808, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.16353857517242432, | |
| "eval_runtime": 22.5783, | |
| "eval_samples_per_second": 1.993, | |
| "eval_steps_per_second": 1.993, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 1.2059634923934937, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.1997, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 1.1025176048278809, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.2105, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 1.010790228843689, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2017, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.8034100532531738, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.1584, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.9222174882888794, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.1708, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 0.6132246851921082, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.1535, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.6803828477859497, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.1848, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.6978905200958252, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.197, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.8006490468978882, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.1337, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.6057296991348267, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.1512, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.11469355970621109, | |
| "eval_runtime": 22.4262, | |
| "eval_samples_per_second": 2.007, | |
| "eval_steps_per_second": 2.007, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.428353488445282, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.102, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.48068785667419434, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.096, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.4888728857040405, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.1044, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.6380906105041504, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.1672, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.6851835250854492, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.1417, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 1.0068365335464478, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.2687, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.45691290497779846, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0871, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.5721675157546997, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.1377, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.47928979992866516, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.0784, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.5411726236343384, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.0879, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.10483992844820023, | |
| "eval_runtime": 22.423, | |
| "eval_samples_per_second": 2.007, | |
| "eval_steps_per_second": 2.007, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.5184138417243958, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.0919, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.5770514011383057, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.121, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.7286047339439392, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.1523, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.574654757976532, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.1063, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.5332781076431274, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.0932, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.5827326774597168, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.0832, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.6699033379554749, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.0958, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 1.0102177858352661, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.1254, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.4694126546382904, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.0807, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 0.5243706107139587, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.096, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.09142621606588364, | |
| "eval_runtime": 22.3949, | |
| "eval_samples_per_second": 2.009, | |
| "eval_steps_per_second": 2.009, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.5594834685325623, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.1036, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.48444709181785583, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.0667, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.555347740650177, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.0981, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.5066335797309875, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.067, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.4888894855976105, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.0882, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.6094079613685608, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.0988, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.7239038944244385, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.0887, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.5987774729728699, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.1288, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.4367763102054596, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.0705, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.4437493681907654, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.0572, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.08621598780155182, | |
| "eval_runtime": 22.4022, | |
| "eval_samples_per_second": 2.009, | |
| "eval_steps_per_second": 2.009, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 0.9511323571205139, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.1278, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.3670876920223236, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.0576, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.46987372636795044, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.0899, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.4614879786968231, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.0873, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.36645394563674927, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.0672, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.3303259313106537, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.0444, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.468173623085022, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.0641, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.5283992886543274, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.1024, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.4989950358867645, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.0465, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.42546793818473816, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.0586, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.08327701687812805, | |
| "eval_runtime": 22.4042, | |
| "eval_samples_per_second": 2.009, | |
| "eval_steps_per_second": 2.009, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.4836946725845337, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.0611, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.5960677266120911, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.0697, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.43261033296585083, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.0612, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.6163232326507568, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.0736, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.4289225935935974, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.0743, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.4335273504257202, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.0549, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.5130767822265625, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.0923, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.29725366830825806, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.0407, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 0.49687743186950684, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0857, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.5076261758804321, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.0825, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.08645560592412949, | |
| "eval_runtime": 22.4305, | |
| "eval_samples_per_second": 2.006, | |
| "eval_steps_per_second": 2.006, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.6246863603591919, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.0623, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.555365264415741, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.0496, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.48801174759864807, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.0619, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.4042954444885254, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.0542, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.4558180570602417, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.0514, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 0.9928022027015686, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.1532, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.29005953669548035, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0372, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.29069948196411133, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.0417, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.34006467461586, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.0557, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.386521577835083, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.0413, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.08211181312799454, | |
| "eval_runtime": 22.425, | |
| "eval_samples_per_second": 2.007, | |
| "eval_steps_per_second": 2.007, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.435297429561615, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.0533, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.42633265256881714, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0595, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.40960946679115295, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.039, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.42120492458343506, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.0422, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.5309680104255676, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0796, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.6197146773338318, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.0575, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.3979469835758209, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.0377, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.4420924186706543, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.0603, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.5277808308601379, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.0686, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.4421432316303253, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.0595, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.08340814709663391, | |
| "eval_runtime": 22.422, | |
| "eval_samples_per_second": 2.007, | |
| "eval_steps_per_second": 2.007, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.4125105142593384, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.0539, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.325244277715683, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.0259, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.3284305930137634, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.0281, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.5049591660499573, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.0559, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.6370705366134644, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.0532, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.412969172000885, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0395, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.3975721597671509, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.045, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.5301607251167297, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.0479, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.3365274667739868, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.032, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.4609011709690094, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.0487, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.08508718758821487, | |
| "eval_runtime": 22.4554, | |
| "eval_samples_per_second": 2.004, | |
| "eval_steps_per_second": 2.004, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 0.561644434928894, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.0686, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 1.3235689401626587, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.1164, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.328523188829422, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.0323, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.32442378997802734, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0245, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.39211609959602356, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0341, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.4840279817581177, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0622, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.40528035163879395, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.0382, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.4375377297401428, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.0418, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.5347009301185608, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.0547, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.5620562434196472, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0374, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.08784408122301102, | |
| "eval_runtime": 22.387, | |
| "eval_samples_per_second": 2.01, | |
| "eval_steps_per_second": 2.01, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.3576813042163849, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0341, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.44771745800971985, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0402, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.3740685284137726, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.0317, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.3905841112136841, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.0368, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.5260807275772095, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0419, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.3988063931465149, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.0511, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.4558522701263428, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.0387, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.4286305606365204, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0532, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.44637882709503174, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.043, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.3593423664569855, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.0293, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.08751612901687622, | |
| "eval_runtime": 22.3888, | |
| "eval_samples_per_second": 2.01, | |
| "eval_steps_per_second": 2.01, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.2957054376602173, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.0178, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.4842686653137207, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0467, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.6364602446556091, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.0646, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.49542567133903503, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0487, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.44883695244789124, | |
| "learning_rate": 0.0, | |
| "loss": 0.046, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.3328024995882598e+17, | |
| "train_loss": 0.09674194711446762, | |
| "train_runtime": 3718.0799, | |
| "train_samples_per_second": 0.545, | |
| "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": 1.3328024995882598e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |