| { | |
| "best_metric": 0.14818251132965088, | |
| "best_model_checkpoint": "saves/psy-course-info/Llama3-OpenBioLLM-8B/train/fold6/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": 3.694704055786133, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.9769, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.07901234567901234, | |
| "grad_norm": 3.414362668991089, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.0967, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.11851851851851852, | |
| "grad_norm": 3.79213547706604, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 0.9042, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1580246913580247, | |
| "grad_norm": 3.0754430294036865, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 1.1026, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 2.763463258743286, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 0.9803, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 3.1616077423095703, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.9816, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.2765432098765432, | |
| "grad_norm": 2.723661422729492, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 0.894, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3160493827160494, | |
| "grad_norm": 1.9742035865783691, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.7963, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 2.1275100708007812, | |
| "learning_rate": 6.923076923076924e-05, | |
| "loss": 0.6038, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 1.6750327348709106, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.5581, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "eval_loss": 0.42407938838005066, | |
| "eval_runtime": 19.1438, | |
| "eval_samples_per_second": 2.351, | |
| "eval_steps_per_second": 2.351, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4345679012345679, | |
| "grad_norm": 1.5146023035049438, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.5228, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 1.1980900764465332, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.434, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5135802469135803, | |
| "grad_norm": 1.000938892364502, | |
| "learning_rate": 0.0001, | |
| "loss": 0.3648, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.5530864197530864, | |
| "grad_norm": 0.9401538372039795, | |
| "learning_rate": 9.998033131915266e-05, | |
| "loss": 0.3034, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.5925925925925926, | |
| "grad_norm": 0.8780136108398438, | |
| "learning_rate": 9.992134075089084e-05, | |
| "loss": 0.3078, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.6320987654320988, | |
| "grad_norm": 1.0185179710388184, | |
| "learning_rate": 9.982307470588098e-05, | |
| "loss": 0.3423, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.671604938271605, | |
| "grad_norm": 0.9036634564399719, | |
| "learning_rate": 9.968561049466214e-05, | |
| "loss": 0.2821, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 0.7214317917823792, | |
| "learning_rate": 9.950905626682228e-05, | |
| "loss": 0.2518, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.7506172839506173, | |
| "grad_norm": 0.6148759722709656, | |
| "learning_rate": 9.92935509259118e-05, | |
| "loss": 0.1671, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "grad_norm": 0.7838930487632751, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 0.2468, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7901234567901234, | |
| "eval_loss": 0.22477315366268158, | |
| "eval_runtime": 19.1265, | |
| "eval_samples_per_second": 2.353, | |
| "eval_steps_per_second": 2.353, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8296296296296296, | |
| "grad_norm": 0.7086945176124573, | |
| "learning_rate": 9.874639560909117e-05, | |
| "loss": 0.2517, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.8691358024691358, | |
| "grad_norm": 0.8672019839286804, | |
| "learning_rate": 9.841517610611309e-05, | |
| "loss": 0.2909, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.908641975308642, | |
| "grad_norm": 0.6496129035949707, | |
| "learning_rate": 9.804586609725499e-05, | |
| "loss": 0.1736, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 0.8389897346496582, | |
| "learning_rate": 9.763875613614482e-05, | |
| "loss": 0.2561, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.9876543209876543, | |
| "grad_norm": 0.7387816905975342, | |
| "learning_rate": 9.719416651541839e-05, | |
| "loss": 0.2135, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0271604938271606, | |
| "grad_norm": 1.6426275968551636, | |
| "learning_rate": 9.671244701472999e-05, | |
| "loss": 0.3146, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.0666666666666667, | |
| "grad_norm": 0.6529827117919922, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.1692, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.106172839506173, | |
| "grad_norm": 0.6806563138961792, | |
| "learning_rate": 9.563916325306594e-05, | |
| "loss": 0.1944, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.145679012345679, | |
| "grad_norm": 0.7545333504676819, | |
| "learning_rate": 9.504844339512095e-05, | |
| "loss": 0.2234, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "grad_norm": 0.6371580362319946, | |
| "learning_rate": 9.442228179894362e-05, | |
| "loss": 0.1806, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1851851851851851, | |
| "eval_loss": 0.18481381237506866, | |
| "eval_runtime": 19.1087, | |
| "eval_samples_per_second": 2.355, | |
| "eval_steps_per_second": 2.355, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2246913580246914, | |
| "grad_norm": 0.9482006430625916, | |
| "learning_rate": 9.376117109543769e-05, | |
| "loss": 0.1917, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.2641975308641975, | |
| "grad_norm": 0.8220361471176147, | |
| "learning_rate": 9.306563141162046e-05, | |
| "loss": 0.1575, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.3037037037037038, | |
| "grad_norm": 0.8422967195510864, | |
| "learning_rate": 9.233620996141421e-05, | |
| "loss": 0.1346, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.34320987654321, | |
| "grad_norm": 0.829232931137085, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 0.1451, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.382716049382716, | |
| "grad_norm": 0.8307307362556458, | |
| "learning_rate": 9.077804344796302e-05, | |
| "loss": 0.1833, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4222222222222223, | |
| "grad_norm": 0.8141887187957764, | |
| "learning_rate": 8.995052426791247e-05, | |
| "loss": 0.1981, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.4617283950617284, | |
| "grad_norm": 0.7708631157875061, | |
| "learning_rate": 8.90915741234015e-05, | |
| "loss": 0.164, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.5012345679012347, | |
| "grad_norm": 0.6389580965042114, | |
| "learning_rate": 8.820186879108038e-05, | |
| "loss": 0.1476, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.5407407407407407, | |
| "grad_norm": 0.7386507987976074, | |
| "learning_rate": 8.728210824415827e-05, | |
| "loss": 0.2438, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "grad_norm": 1.0356231927871704, | |
| "learning_rate": 8.633301610170135e-05, | |
| "loss": 0.2529, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5802469135802468, | |
| "eval_loss": 0.17059147357940674, | |
| "eval_runtime": 19.112, | |
| "eval_samples_per_second": 2.355, | |
| "eval_steps_per_second": 2.355, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6197530864197531, | |
| "grad_norm": 0.6339937448501587, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.1273, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.6592592592592592, | |
| "grad_norm": 0.7589173913002014, | |
| "learning_rate": 8.434984630174509e-05, | |
| "loss": 0.1275, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.6987654320987655, | |
| "grad_norm": 0.6136049032211304, | |
| "learning_rate": 8.33173288976002e-05, | |
| "loss": 0.1485, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.7382716049382716, | |
| "grad_norm": 0.6310786008834839, | |
| "learning_rate": 8.225859917710439e-05, | |
| "loss": 0.0977, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.7777777777777777, | |
| "grad_norm": 0.6631302237510681, | |
| "learning_rate": 8.117449009293668e-05, | |
| "loss": 0.1359, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.817283950617284, | |
| "grad_norm": 0.6856066584587097, | |
| "learning_rate": 8.006585456492029e-05, | |
| "loss": 0.1586, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.8567901234567903, | |
| "grad_norm": 0.5605504512786865, | |
| "learning_rate": 7.89335648089903e-05, | |
| "loss": 0.1261, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.8962962962962964, | |
| "grad_norm": 0.8561685085296631, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 0.1741, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.9358024691358025, | |
| "grad_norm": 0.8282690048217773, | |
| "learning_rate": 7.660160382576683e-05, | |
| "loss": 0.179, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "grad_norm": 0.569130003452301, | |
| "learning_rate": 7.540376726232648e-05, | |
| "loss": 0.1665, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.9753086419753085, | |
| "eval_loss": 0.15281587839126587, | |
| "eval_runtime": 19.0967, | |
| "eval_samples_per_second": 2.356, | |
| "eval_steps_per_second": 2.356, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0148148148148146, | |
| "grad_norm": 1.3159841299057007, | |
| "learning_rate": 7.4185944355262e-05, | |
| "loss": 0.2569, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.054320987654321, | |
| "grad_norm": 0.3862403333187103, | |
| "learning_rate": 7.294909322337689e-05, | |
| "loss": 0.0956, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.093827160493827, | |
| "grad_norm": 0.4631720185279846, | |
| "learning_rate": 7.169418695587791e-05, | |
| "loss": 0.1025, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.1333333333333333, | |
| "grad_norm": 0.5311153531074524, | |
| "learning_rate": 7.042221284679982e-05, | |
| "loss": 0.1335, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.1728395061728394, | |
| "grad_norm": 0.54117751121521, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1441, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.212345679012346, | |
| "grad_norm": 0.604268491268158, | |
| "learning_rate": 6.783107663311565e-05, | |
| "loss": 0.1312, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.251851851851852, | |
| "grad_norm": 0.6387006640434265, | |
| "learning_rate": 6.651395309775837e-05, | |
| "loss": 0.1058, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.291358024691358, | |
| "grad_norm": 0.6895653009414673, | |
| "learning_rate": 6.518383725548074e-05, | |
| "loss": 0.1457, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.330864197530864, | |
| "grad_norm": 0.4936845600605011, | |
| "learning_rate": 6.384177557124247e-05, | |
| "loss": 0.0918, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "grad_norm": 0.6826022267341614, | |
| "learning_rate": 6.248882390836135e-05, | |
| "loss": 0.1308, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.3703703703703702, | |
| "eval_loss": 0.15456868708133698, | |
| "eval_runtime": 19.1185, | |
| "eval_samples_per_second": 2.354, | |
| "eval_steps_per_second": 2.354, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.4098765432098768, | |
| "grad_norm": 0.736090362071991, | |
| "learning_rate": 6.112604669781572e-05, | |
| "loss": 0.0988, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.449382716049383, | |
| "grad_norm": 0.6230776906013489, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 0.0819, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.488888888888889, | |
| "grad_norm": 0.8704579472541809, | |
| "learning_rate": 5.837531116523682e-05, | |
| "loss": 0.1146, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.528395061728395, | |
| "grad_norm": 0.8533468842506409, | |
| "learning_rate": 5.698951697677498e-05, | |
| "loss": 0.0864, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.567901234567901, | |
| "grad_norm": 0.7294342517852783, | |
| "learning_rate": 5.559822380516539e-05, | |
| "loss": 0.1177, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.6074074074074076, | |
| "grad_norm": 0.7356113195419312, | |
| "learning_rate": 5.420252624646238e-05, | |
| "loss": 0.1538, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.6469135802469137, | |
| "grad_norm": 0.8724901080131531, | |
| "learning_rate": 5.2803522361859594e-05, | |
| "loss": 0.1298, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.68641975308642, | |
| "grad_norm": 0.6292612552642822, | |
| "learning_rate": 5.140231281379345e-05, | |
| "loss": 0.0819, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 2.725925925925926, | |
| "grad_norm": 1.1373646259307861, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1255, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "grad_norm": 0.6750616431236267, | |
| "learning_rate": 4.859768718620656e-05, | |
| "loss": 0.1003, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.765432098765432, | |
| "eval_loss": 0.14987002313137054, | |
| "eval_runtime": 19.118, | |
| "eval_samples_per_second": 2.354, | |
| "eval_steps_per_second": 2.354, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.8049382716049385, | |
| "grad_norm": 0.8765468597412109, | |
| "learning_rate": 4.7196477638140404e-05, | |
| "loss": 0.1024, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.8444444444444446, | |
| "grad_norm": 0.6719949841499329, | |
| "learning_rate": 4.579747375353763e-05, | |
| "loss": 0.0993, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.8839506172839506, | |
| "grad_norm": 0.6172820925712585, | |
| "learning_rate": 4.4401776194834613e-05, | |
| "loss": 0.0831, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.9234567901234567, | |
| "grad_norm": 0.8515627384185791, | |
| "learning_rate": 4.3010483023225045e-05, | |
| "loss": 0.0974, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.962962962962963, | |
| "grad_norm": 0.7816348075866699, | |
| "learning_rate": 4.162468883476319e-05, | |
| "loss": 0.0924, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0024691358024693, | |
| "grad_norm": 1.9471156597137451, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 0.236, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.0419753086419754, | |
| "grad_norm": 0.7431677579879761, | |
| "learning_rate": 3.887395330218429e-05, | |
| "loss": 0.0828, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.0814814814814815, | |
| "grad_norm": 0.5865127444267273, | |
| "learning_rate": 3.7511176091638653e-05, | |
| "loss": 0.0712, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.1209876543209876, | |
| "grad_norm": 0.5857524275779724, | |
| "learning_rate": 3.6158224428757535e-05, | |
| "loss": 0.0946, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "grad_norm": 0.5269191265106201, | |
| "learning_rate": 3.4816162744519263e-05, | |
| "loss": 0.0591, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.1604938271604937, | |
| "eval_loss": 0.14818251132965088, | |
| "eval_runtime": 19.093, | |
| "eval_samples_per_second": 2.357, | |
| "eval_steps_per_second": 2.357, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 0.734276294708252, | |
| "learning_rate": 3.3486046902241664e-05, | |
| "loss": 0.0626, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.2395061728395063, | |
| "grad_norm": 0.677424430847168, | |
| "learning_rate": 3.216892336688435e-05, | |
| "loss": 0.0877, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.2790123456790123, | |
| "grad_norm": 0.74637770652771, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0953, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.3185185185185184, | |
| "grad_norm": 0.6536086797714233, | |
| "learning_rate": 2.9577787153200197e-05, | |
| "loss": 0.0886, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.3580246913580245, | |
| "grad_norm": 0.5924112796783447, | |
| "learning_rate": 2.8305813044122097e-05, | |
| "loss": 0.0593, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.397530864197531, | |
| "grad_norm": 0.635668933391571, | |
| "learning_rate": 2.705090677662311e-05, | |
| "loss": 0.0839, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.437037037037037, | |
| "grad_norm": 0.7729941010475159, | |
| "learning_rate": 2.581405564473801e-05, | |
| "loss": 0.09, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.476543209876543, | |
| "grad_norm": 0.5171845555305481, | |
| "learning_rate": 2.459623273767354e-05, | |
| "loss": 0.0819, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.5160493827160493, | |
| "grad_norm": 0.6740212440490723, | |
| "learning_rate": 2.3398396174233178e-05, | |
| "loss": 0.0875, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "grad_norm": 0.6716038584709167, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 0.1073, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.5555555555555554, | |
| "eval_loss": 0.15476737916469574, | |
| "eval_runtime": 19.1485, | |
| "eval_samples_per_second": 2.35, | |
| "eval_steps_per_second": 2.35, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.595061728395062, | |
| "grad_norm": 0.6998808979988098, | |
| "learning_rate": 2.1066435191009715e-05, | |
| "loss": 0.0935, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 3.634567901234568, | |
| "grad_norm": 0.7683432698249817, | |
| "learning_rate": 1.9934145435079702e-05, | |
| "loss": 0.0922, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 3.674074074074074, | |
| "grad_norm": 0.45504406094551086, | |
| "learning_rate": 1.8825509907063327e-05, | |
| "loss": 0.0414, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 3.71358024691358, | |
| "grad_norm": 0.606503963470459, | |
| "learning_rate": 1.774140082289563e-05, | |
| "loss": 0.0866, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 3.753086419753086, | |
| "grad_norm": 0.5964378118515015, | |
| "learning_rate": 1.6682671102399805e-05, | |
| "loss": 0.0902, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.7925925925925927, | |
| "grad_norm": 0.6444228291511536, | |
| "learning_rate": 1.5650153698254916e-05, | |
| "loss": 0.0565, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 3.832098765432099, | |
| "grad_norm": 0.7363929152488708, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.093, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 3.871604938271605, | |
| "grad_norm": 0.8473097085952759, | |
| "learning_rate": 1.3666983898298657e-05, | |
| "loss": 0.1005, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 3.911111111111111, | |
| "grad_norm": 0.6378225684165955, | |
| "learning_rate": 1.2717891755841722e-05, | |
| "loss": 0.0907, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "grad_norm": 0.6959088444709778, | |
| "learning_rate": 1.1798131208919627e-05, | |
| "loss": 0.1003, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.950617283950617, | |
| "eval_loss": 0.14895030856132507, | |
| "eval_runtime": 19.1622, | |
| "eval_samples_per_second": 2.348, | |
| "eval_steps_per_second": 2.348, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.9901234567901236, | |
| "grad_norm": 1.1908460855484009, | |
| "learning_rate": 1.090842587659851e-05, | |
| "loss": 0.1074, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.029629629629629, | |
| "grad_norm": 0.9052301049232483, | |
| "learning_rate": 1.004947573208756e-05, | |
| "loss": 0.1072, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.069135802469136, | |
| "grad_norm": 0.5730845332145691, | |
| "learning_rate": 9.221956552036992e-06, | |
| "loss": 0.0667, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.108641975308642, | |
| "grad_norm": 0.42981043457984924, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 0.0373, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.148148148148148, | |
| "grad_norm": 0.5426796078681946, | |
| "learning_rate": 7.663790038585793e-06, | |
| "loss": 0.0799, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.187654320987654, | |
| "grad_norm": 0.707223653793335, | |
| "learning_rate": 6.934368588379553e-06, | |
| "loss": 0.0726, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.22716049382716, | |
| "grad_norm": 0.6215255856513977, | |
| "learning_rate": 6.238828904562316e-06, | |
| "loss": 0.0664, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.266666666666667, | |
| "grad_norm": 0.6359230875968933, | |
| "learning_rate": 5.577718201056392e-06, | |
| "loss": 0.0588, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.306172839506173, | |
| "grad_norm": 0.5783518552780151, | |
| "learning_rate": 4.951556604879048e-06, | |
| "loss": 0.0636, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "grad_norm": 0.4861449599266052, | |
| "learning_rate": 4.360836746934055e-06, | |
| "loss": 0.0791, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.345679012345679, | |
| "eval_loss": 0.15216168761253357, | |
| "eval_runtime": 19.1654, | |
| "eval_samples_per_second": 2.348, | |
| "eval_steps_per_second": 2.348, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.385185185185185, | |
| "grad_norm": 0.5029930472373962, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.049, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.424691358024692, | |
| "grad_norm": 0.6033341884613037, | |
| "learning_rate": 3.2875529852700147e-06, | |
| "loss": 0.0855, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.4641975308641975, | |
| "grad_norm": 0.4884686768054962, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.0786, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.503703703703704, | |
| "grad_norm": 0.4735491871833801, | |
| "learning_rate": 2.361243863855184e-06, | |
| "loss": 0.0627, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 4.54320987654321, | |
| "grad_norm": 0.4757883846759796, | |
| "learning_rate": 1.9541339027450256e-06, | |
| "loss": 0.0591, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.582716049382716, | |
| "grad_norm": 0.5591762065887451, | |
| "learning_rate": 1.584823893886933e-06, | |
| "loss": 0.0738, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 4.622222222222222, | |
| "grad_norm": 0.5646064281463623, | |
| "learning_rate": 1.2536043909088191e-06, | |
| "loss": 0.0613, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 4.661728395061728, | |
| "grad_norm": 0.6605923175811768, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 0.0892, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 4.701234567901235, | |
| "grad_norm": 0.4959077537059784, | |
| "learning_rate": 7.064490740882057e-07, | |
| "loss": 0.0592, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "grad_norm": 0.4682895541191101, | |
| "learning_rate": 4.909437331777179e-07, | |
| "loss": 0.0557, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.7407407407407405, | |
| "eval_loss": 0.15291602909564972, | |
| "eval_runtime": 19.2034, | |
| "eval_samples_per_second": 2.343, | |
| "eval_steps_per_second": 2.343, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.780246913580247, | |
| "grad_norm": 0.5253302454948425, | |
| "learning_rate": 3.143895053378698e-07, | |
| "loss": 0.0567, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 4.8197530864197535, | |
| "grad_norm": 0.6253769397735596, | |
| "learning_rate": 1.7692529411904578e-07, | |
| "loss": 0.0687, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 4.859259259259259, | |
| "grad_norm": 0.634563684463501, | |
| "learning_rate": 7.865924910916977e-08, | |
| "loss": 0.0844, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 4.898765432098766, | |
| "grad_norm": 0.5740000009536743, | |
| "learning_rate": 1.9668680847356735e-08, | |
| "loss": 0.0896, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "grad_norm": 0.4100923538208008, | |
| "learning_rate": 0.0, | |
| "loss": 0.0472, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.938271604938271, | |
| "step": 125, | |
| "total_flos": 1.2820912048314778e+17, | |
| "train_loss": 0.19632812148332596, | |
| "train_runtime": 3461.189, | |
| "train_samples_per_second": 0.585, | |
| "train_steps_per_second": 0.036 | |
| } | |
| ], | |
| "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": 1.2820912048314778e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |