| { |
| "best_global_step": 399, |
| "best_metric": 0.13073493540287018, |
| "best_model_checkpoint": "saves_multiple/p-tuning/llama-3-8b-instruct/train_cb_123_1760637639/checkpoint-399", |
| "epoch": 20.0, |
| "eval_steps": 57, |
| "global_step": 1140, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.08771929824561403, |
| "grad_norm": 128.95028686523438, |
| "learning_rate": 3.508771929824561e-05, |
| "loss": 8.5412, |
| "num_input_tokens_seen": 3552, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.17543859649122806, |
| "grad_norm": 61.61833190917969, |
| "learning_rate": 7.894736842105263e-05, |
| "loss": 2.8916, |
| "num_input_tokens_seen": 7264, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.2631578947368421, |
| "grad_norm": 59.73617935180664, |
| "learning_rate": 0.00012280701754385965, |
| "loss": 0.6364, |
| "num_input_tokens_seen": 10528, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.3508771929824561, |
| "grad_norm": 37.573612213134766, |
| "learning_rate": 0.00016666666666666666, |
| "loss": 0.8324, |
| "num_input_tokens_seen": 14784, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.43859649122807015, |
| "grad_norm": 35.550758361816406, |
| "learning_rate": 0.00021052631578947367, |
| "loss": 0.4937, |
| "num_input_tokens_seen": 18112, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.5263157894736842, |
| "grad_norm": 28.5069580078125, |
| "learning_rate": 0.0002543859649122807, |
| "loss": 1.8405, |
| "num_input_tokens_seen": 20736, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6140350877192983, |
| "grad_norm": 37.91823196411133, |
| "learning_rate": 0.0002982456140350877, |
| "loss": 0.5002, |
| "num_input_tokens_seen": 24896, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.7017543859649122, |
| "grad_norm": 4.9007887840271, |
| "learning_rate": 0.00034210526315789477, |
| "loss": 0.2443, |
| "num_input_tokens_seen": 28160, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.7894736842105263, |
| "grad_norm": 2.743189811706543, |
| "learning_rate": 0.00038596491228070175, |
| "loss": 0.2807, |
| "num_input_tokens_seen": 31040, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.8771929824561403, |
| "grad_norm": 3.108569860458374, |
| "learning_rate": 0.0004298245614035088, |
| "loss": 0.3373, |
| "num_input_tokens_seen": 33760, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.9649122807017544, |
| "grad_norm": 1.4290847778320312, |
| "learning_rate": 0.00047368421052631577, |
| "loss": 0.2819, |
| "num_input_tokens_seen": 36416, |
| "step": 55 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.758574903011322, |
| "eval_runtime": 0.7377, |
| "eval_samples_per_second": 33.889, |
| "eval_steps_per_second": 9.489, |
| "num_input_tokens_seen": 37160, |
| "step": 57 |
| }, |
| { |
| "epoch": 1.0526315789473684, |
| "grad_norm": 7.512516975402832, |
| "learning_rate": 0.0005175438596491229, |
| "loss": 0.221, |
| "num_input_tokens_seen": 39176, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.1403508771929824, |
| "grad_norm": 16.071565628051758, |
| "learning_rate": 0.0005614035087719298, |
| "loss": 0.358, |
| "num_input_tokens_seen": 42632, |
| "step": 65 |
| }, |
| { |
| "epoch": 1.2280701754385965, |
| "grad_norm": 2.7750377655029297, |
| "learning_rate": 0.0006052631578947369, |
| "loss": 0.3661, |
| "num_input_tokens_seen": 45704, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.3157894736842106, |
| "grad_norm": 93.76712036132812, |
| "learning_rate": 0.0006491228070175439, |
| "loss": 0.3337, |
| "num_input_tokens_seen": 49448, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.4035087719298245, |
| "grad_norm": 8.643919944763184, |
| "learning_rate": 0.0006929824561403509, |
| "loss": 0.7437, |
| "num_input_tokens_seen": 52456, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.4912280701754386, |
| "grad_norm": 1.2434449195861816, |
| "learning_rate": 0.0007368421052631579, |
| "loss": 0.2926, |
| "num_input_tokens_seen": 56488, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.5789473684210527, |
| "grad_norm": 2.895885944366455, |
| "learning_rate": 0.0007807017543859649, |
| "loss": 0.4001, |
| "num_input_tokens_seen": 59208, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.6666666666666665, |
| "grad_norm": 1.0412471294403076, |
| "learning_rate": 0.000824561403508772, |
| "loss": 0.1042, |
| "num_input_tokens_seen": 62696, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.7543859649122808, |
| "grad_norm": 2.8564751148223877, |
| "learning_rate": 0.000868421052631579, |
| "loss": 0.1886, |
| "num_input_tokens_seen": 66024, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.8421052631578947, |
| "grad_norm": 1.2726372480392456, |
| "learning_rate": 0.000912280701754386, |
| "loss": 0.3836, |
| "num_input_tokens_seen": 69000, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.9298245614035088, |
| "grad_norm": 0.24507200717926025, |
| "learning_rate": 0.0009561403508771929, |
| "loss": 0.1652, |
| "num_input_tokens_seen": 72072, |
| "step": 110 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.21064288914203644, |
| "eval_runtime": 0.744, |
| "eval_samples_per_second": 33.604, |
| "eval_steps_per_second": 9.409, |
| "num_input_tokens_seen": 73720, |
| "step": 114 |
| }, |
| { |
| "epoch": 2.017543859649123, |
| "grad_norm": 0.7454385161399841, |
| "learning_rate": 0.001, |
| "loss": 0.0736, |
| "num_input_tokens_seen": 74264, |
| "step": 115 |
| }, |
| { |
| "epoch": 2.1052631578947367, |
| "grad_norm": 0.21397805213928223, |
| "learning_rate": 0.000999941402841295, |
| "loss": 0.0278, |
| "num_input_tokens_seen": 77528, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.192982456140351, |
| "grad_norm": 0.030098894611001015, |
| "learning_rate": 0.0009997656250996883, |
| "loss": 0.0966, |
| "num_input_tokens_seen": 80600, |
| "step": 125 |
| }, |
| { |
| "epoch": 2.280701754385965, |
| "grad_norm": 1.9476398229599, |
| "learning_rate": 0.0009994727079754844, |
| "loss": 0.2501, |
| "num_input_tokens_seen": 83800, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.3684210526315788, |
| "grad_norm": 0.6966570019721985, |
| "learning_rate": 0.0009990627201251284, |
| "loss": 0.3148, |
| "num_input_tokens_seen": 86552, |
| "step": 135 |
| }, |
| { |
| "epoch": 2.456140350877193, |
| "grad_norm": 0.9255178570747375, |
| "learning_rate": 0.0009985357576451127, |
| "loss": 0.0624, |
| "num_input_tokens_seen": 90008, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.543859649122807, |
| "grad_norm": 0.1348254233598709, |
| "learning_rate": 0.0009978919440494537, |
| "loss": 0.4121, |
| "num_input_tokens_seen": 93496, |
| "step": 145 |
| }, |
| { |
| "epoch": 2.6315789473684212, |
| "grad_norm": 0.49739935994148254, |
| "learning_rate": 0.0009971314302407413, |
| "loss": 0.0957, |
| "num_input_tokens_seen": 96888, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.719298245614035, |
| "grad_norm": 0.5351743102073669, |
| "learning_rate": 0.0009962543944747686, |
| "loss": 0.0688, |
| "num_input_tokens_seen": 100728, |
| "step": 155 |
| }, |
| { |
| "epoch": 2.807017543859649, |
| "grad_norm": 0.04349172115325928, |
| "learning_rate": 0.0009952610423187517, |
| "loss": 0.1149, |
| "num_input_tokens_seen": 103960, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.8947368421052633, |
| "grad_norm": 0.695615828037262, |
| "learning_rate": 0.0009941516066031462, |
| "loss": 0.159, |
| "num_input_tokens_seen": 106648, |
| "step": 165 |
| }, |
| { |
| "epoch": 2.982456140350877, |
| "grad_norm": 0.04241182282567024, |
| "learning_rate": 0.0009929263473670749, |
| "loss": 0.1091, |
| "num_input_tokens_seen": 110136, |
| "step": 170 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.16814041137695312, |
| "eval_runtime": 0.7418, |
| "eval_samples_per_second": 33.701, |
| "eval_steps_per_second": 9.436, |
| "num_input_tokens_seen": 110296, |
| "step": 171 |
| }, |
| { |
| "epoch": 3.0701754385964914, |
| "grad_norm": 0.3171384930610657, |
| "learning_rate": 0.0009915855517973774, |
| "loss": 0.0703, |
| "num_input_tokens_seen": 112984, |
| "step": 175 |
| }, |
| { |
| "epoch": 3.1578947368421053, |
| "grad_norm": 0.40988221764564514, |
| "learning_rate": 0.0009901295341612972, |
| "loss": 0.0453, |
| "num_input_tokens_seen": 116120, |
| "step": 180 |
| }, |
| { |
| "epoch": 3.245614035087719, |
| "grad_norm": 0.5177841782569885, |
| "learning_rate": 0.00098855863573282, |
| "loss": 0.1497, |
| "num_input_tokens_seen": 119576, |
| "step": 185 |
| }, |
| { |
| "epoch": 3.3333333333333335, |
| "grad_norm": 0.495584100484848, |
| "learning_rate": 0.0009868732247126839, |
| "loss": 0.0167, |
| "num_input_tokens_seen": 122968, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.4210526315789473, |
| "grad_norm": 0.038633063435554504, |
| "learning_rate": 0.000985073696142077, |
| "loss": 0.1827, |
| "num_input_tokens_seen": 126104, |
| "step": 195 |
| }, |
| { |
| "epoch": 3.5087719298245617, |
| "grad_norm": 0.3946753442287445, |
| "learning_rate": 0.0009831604718100442, |
| "loss": 0.2164, |
| "num_input_tokens_seen": 129368, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.5964912280701755, |
| "grad_norm": 0.38737550377845764, |
| "learning_rate": 0.0009811340001546253, |
| "loss": 0.0977, |
| "num_input_tokens_seen": 132888, |
| "step": 205 |
| }, |
| { |
| "epoch": 3.6842105263157894, |
| "grad_norm": 0.36464640498161316, |
| "learning_rate": 0.0009789947561577445, |
| "loss": 0.06, |
| "num_input_tokens_seen": 135832, |
| "step": 210 |
| }, |
| { |
| "epoch": 3.7719298245614032, |
| "grad_norm": 0.11651939898729324, |
| "learning_rate": 0.000976743241233882, |
| "loss": 0.1588, |
| "num_input_tokens_seen": 138936, |
| "step": 215 |
| }, |
| { |
| "epoch": 3.8596491228070176, |
| "grad_norm": 0.17754873633384705, |
| "learning_rate": 0.0009743799831125471, |
| "loss": 0.0407, |
| "num_input_tokens_seen": 142488, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.9473684210526314, |
| "grad_norm": 0.3502120077610016, |
| "learning_rate": 0.0009719055357145847, |
| "loss": 0.0807, |
| "num_input_tokens_seen": 146808, |
| "step": 225 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.18574421107769012, |
| "eval_runtime": 0.7462, |
| "eval_samples_per_second": 33.501, |
| "eval_steps_per_second": 9.38, |
| "num_input_tokens_seen": 147784, |
| "step": 228 |
| }, |
| { |
| "epoch": 4.035087719298246, |
| "grad_norm": 0.21568936109542847, |
| "learning_rate": 0.0009693204790223423, |
| "loss": 0.0834, |
| "num_input_tokens_seen": 149448, |
| "step": 230 |
| }, |
| { |
| "epoch": 4.12280701754386, |
| "grad_norm": 0.4902511239051819, |
| "learning_rate": 0.0009666254189437286, |
| "loss": 0.0547, |
| "num_input_tokens_seen": 152488, |
| "step": 235 |
| }, |
| { |
| "epoch": 4.2105263157894735, |
| "grad_norm": 1.5583466291427612, |
| "learning_rate": 0.0009638209871701966, |
| "loss": 0.1172, |
| "num_input_tokens_seen": 155208, |
| "step": 240 |
| }, |
| { |
| "epoch": 4.298245614035087, |
| "grad_norm": 0.3295286297798157, |
| "learning_rate": 0.0009609078410286809, |
| "loss": 0.0856, |
| "num_input_tokens_seen": 158088, |
| "step": 245 |
| }, |
| { |
| "epoch": 4.385964912280702, |
| "grad_norm": 0.984485924243927, |
| "learning_rate": 0.0009578866633275287, |
| "loss": 0.1089, |
| "num_input_tokens_seen": 161032, |
| "step": 250 |
| }, |
| { |
| "epoch": 4.473684210526316, |
| "grad_norm": 0.12086866050958633, |
| "learning_rate": 0.0009547581621964571, |
| "loss": 0.0217, |
| "num_input_tokens_seen": 164744, |
| "step": 255 |
| }, |
| { |
| "epoch": 4.56140350877193, |
| "grad_norm": 0.07994663715362549, |
| "learning_rate": 0.0009515230709205749, |
| "loss": 0.0154, |
| "num_input_tokens_seen": 167592, |
| "step": 260 |
| }, |
| { |
| "epoch": 4.649122807017544, |
| "grad_norm": 0.022505775094032288, |
| "learning_rate": 0.0009481821477685101, |
| "loss": 0.0668, |
| "num_input_tokens_seen": 170696, |
| "step": 265 |
| }, |
| { |
| "epoch": 4.7368421052631575, |
| "grad_norm": 0.6926897764205933, |
| "learning_rate": 0.0009447361758146791, |
| "loss": 0.0471, |
| "num_input_tokens_seen": 174024, |
| "step": 270 |
| }, |
| { |
| "epoch": 4.824561403508772, |
| "grad_norm": 0.02440646104514599, |
| "learning_rate": 0.0009411859627557439, |
| "loss": 0.1644, |
| "num_input_tokens_seen": 178152, |
| "step": 275 |
| }, |
| { |
| "epoch": 4.912280701754386, |
| "grad_norm": 0.37806436419487, |
| "learning_rate": 0.0009375323407212969, |
| "loss": 0.0431, |
| "num_input_tokens_seen": 181768, |
| "step": 280 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.04274319112300873, |
| "learning_rate": 0.0009337761660788185, |
| "loss": 0.0258, |
| "num_input_tokens_seen": 184368, |
| "step": 285 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.16147072613239288, |
| "eval_runtime": 0.7468, |
| "eval_samples_per_second": 33.476, |
| "eval_steps_per_second": 9.373, |
| "num_input_tokens_seen": 184368, |
| "step": 285 |
| }, |
| { |
| "epoch": 5.087719298245614, |
| "grad_norm": 0.32277238368988037, |
| "learning_rate": 0.0009299183192329556, |
| "loss": 0.0808, |
| "num_input_tokens_seen": 187312, |
| "step": 290 |
| }, |
| { |
| "epoch": 5.175438596491228, |
| "grad_norm": 0.4225824177265167, |
| "learning_rate": 0.0009259597044191636, |
| "loss": 0.0546, |
| "num_input_tokens_seen": 190736, |
| "step": 295 |
| }, |
| { |
| "epoch": 5.2631578947368425, |
| "grad_norm": 0.3628942668437958, |
| "learning_rate": 0.0009219012494917644, |
| "loss": 0.0442, |
| "num_input_tokens_seen": 194480, |
| "step": 300 |
| }, |
| { |
| "epoch": 5.350877192982456, |
| "grad_norm": 1.066184163093567, |
| "learning_rate": 0.0009177439057064682, |
| "loss": 0.0565, |
| "num_input_tokens_seen": 198576, |
| "step": 305 |
| }, |
| { |
| "epoch": 5.43859649122807, |
| "grad_norm": 0.0820930004119873, |
| "learning_rate": 0.0009134886474974092, |
| "loss": 0.0208, |
| "num_input_tokens_seen": 201808, |
| "step": 310 |
| }, |
| { |
| "epoch": 5.526315789473684, |
| "grad_norm": 0.011820897459983826, |
| "learning_rate": 0.0009091364722487496, |
| "loss": 0.0125, |
| "num_input_tokens_seen": 204912, |
| "step": 315 |
| }, |
| { |
| "epoch": 5.614035087719298, |
| "grad_norm": 0.4663783609867096, |
| "learning_rate": 0.0009046884000609047, |
| "loss": 0.1538, |
| "num_input_tokens_seen": 207760, |
| "step": 320 |
| }, |
| { |
| "epoch": 5.701754385964913, |
| "grad_norm": 0.2975216209888458, |
| "learning_rate": 0.0009001454735114421, |
| "loss": 0.0776, |
| "num_input_tokens_seen": 210544, |
| "step": 325 |
| }, |
| { |
| "epoch": 5.7894736842105265, |
| "grad_norm": 0.3130631148815155, |
| "learning_rate": 0.0008955087574107137, |
| "loss": 0.0804, |
| "num_input_tokens_seen": 214480, |
| "step": 330 |
| }, |
| { |
| "epoch": 5.87719298245614, |
| "grad_norm": 0.6384274363517761, |
| "learning_rate": 0.0008907793385522767, |
| "loss": 0.0508, |
| "num_input_tokens_seen": 218064, |
| "step": 335 |
| }, |
| { |
| "epoch": 5.964912280701754, |
| "grad_norm": 0.5133721232414246, |
| "learning_rate": 0.0008859583254581605, |
| "loss": 0.0447, |
| "num_input_tokens_seen": 220848, |
| "step": 340 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.2889024615287781, |
| "eval_runtime": 0.8611, |
| "eval_samples_per_second": 29.032, |
| "eval_steps_per_second": 8.129, |
| "num_input_tokens_seen": 221536, |
| "step": 342 |
| }, |
| { |
| "epoch": 6.052631578947368, |
| "grad_norm": 0.2570275664329529, |
| "learning_rate": 0.0008810468481190428, |
| "loss": 0.0271, |
| "num_input_tokens_seen": 223616, |
| "step": 345 |
| }, |
| { |
| "epoch": 6.140350877192983, |
| "grad_norm": 0.581085741519928, |
| "learning_rate": 0.000876046057729392, |
| "loss": 0.1896, |
| "num_input_tokens_seen": 227008, |
| "step": 350 |
| }, |
| { |
| "epoch": 6.228070175438597, |
| "grad_norm": 0.4784376621246338, |
| "learning_rate": 0.0008709571264176408, |
| "loss": 0.0695, |
| "num_input_tokens_seen": 230752, |
| "step": 355 |
| }, |
| { |
| "epoch": 6.315789473684211, |
| "grad_norm": 0.12153349071741104, |
| "learning_rate": 0.0008657812469714519, |
| "loss": 0.113, |
| "num_input_tokens_seen": 234208, |
| "step": 360 |
| }, |
| { |
| "epoch": 6.4035087719298245, |
| "grad_norm": 0.22292231023311615, |
| "learning_rate": 0.0008605196325581425, |
| "loss": 0.0643, |
| "num_input_tokens_seen": 237664, |
| "step": 365 |
| }, |
| { |
| "epoch": 6.491228070175438, |
| "grad_norm": 0.36083993315696716, |
| "learning_rate": 0.000855173516440332, |
| "loss": 0.0325, |
| "num_input_tokens_seen": 240704, |
| "step": 370 |
| }, |
| { |
| "epoch": 6.578947368421053, |
| "grad_norm": 0.6128056049346924, |
| "learning_rate": 0.000849744151686879, |
| "loss": 0.1093, |
| "num_input_tokens_seen": 244256, |
| "step": 375 |
| }, |
| { |
| "epoch": 6.666666666666667, |
| "grad_norm": 0.007612931076437235, |
| "learning_rate": 0.000844232810879176, |
| "loss": 0.0406, |
| "num_input_tokens_seen": 246816, |
| "step": 380 |
| }, |
| { |
| "epoch": 6.754385964912281, |
| "grad_norm": 0.006274615880101919, |
| "learning_rate": 0.0008386407858128706, |
| "loss": 0.0096, |
| "num_input_tokens_seen": 249728, |
| "step": 385 |
| }, |
| { |
| "epoch": 6.842105263157895, |
| "grad_norm": 0.02568778023123741, |
| "learning_rate": 0.0008329693871950843, |
| "loss": 0.0167, |
| "num_input_tokens_seen": 252480, |
| "step": 390 |
| }, |
| { |
| "epoch": 6.9298245614035086, |
| "grad_norm": 1.2652301788330078, |
| "learning_rate": 0.0008272199443371966, |
| "loss": 0.0144, |
| "num_input_tokens_seen": 256480, |
| "step": 395 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.13073493540287018, |
| "eval_runtime": 0.7527, |
| "eval_samples_per_second": 33.214, |
| "eval_steps_per_second": 9.3, |
| "num_input_tokens_seen": 258720, |
| "step": 399 |
| }, |
| { |
| "epoch": 7.017543859649122, |
| "grad_norm": 0.002408112632110715, |
| "learning_rate": 0.0008213938048432696, |
| "loss": 0.0027, |
| "num_input_tokens_seen": 259424, |
| "step": 400 |
| }, |
| { |
| "epoch": 7.105263157894737, |
| "grad_norm": 0.003994388971477747, |
| "learning_rate": 0.0008154923342941862, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 262656, |
| "step": 405 |
| }, |
| { |
| "epoch": 7.192982456140351, |
| "grad_norm": 0.017323745414614677, |
| "learning_rate": 0.0008095169159275712, |
| "loss": 0.033, |
| "num_input_tokens_seen": 265408, |
| "step": 410 |
| }, |
| { |
| "epoch": 7.280701754385965, |
| "grad_norm": 1.1956137418746948, |
| "learning_rate": 0.0008034689503135784, |
| "loss": 0.0709, |
| "num_input_tokens_seen": 268992, |
| "step": 415 |
| }, |
| { |
| "epoch": 7.368421052631579, |
| "grad_norm": 0.2515873312950134, |
| "learning_rate": 0.0007973498550266114, |
| "loss": 0.0945, |
| "num_input_tokens_seen": 272288, |
| "step": 420 |
| }, |
| { |
| "epoch": 7.456140350877193, |
| "grad_norm": 0.11299803107976913, |
| "learning_rate": 0.0007911610643130608, |
| "loss": 0.0127, |
| "num_input_tokens_seen": 275104, |
| "step": 425 |
| }, |
| { |
| "epoch": 7.543859649122807, |
| "grad_norm": 0.057308122515678406, |
| "learning_rate": 0.0007849040287551332, |
| "loss": 0.072, |
| "num_input_tokens_seen": 278272, |
| "step": 430 |
| }, |
| { |
| "epoch": 7.631578947368421, |
| "grad_norm": 0.02781853824853897, |
| "learning_rate": 0.000778580214930851, |
| "loss": 0.0749, |
| "num_input_tokens_seen": 282048, |
| "step": 435 |
| }, |
| { |
| "epoch": 7.719298245614035, |
| "grad_norm": 1.6715880632400513, |
| "learning_rate": 0.0007721911050703032, |
| "loss": 0.032, |
| "num_input_tokens_seen": 285536, |
| "step": 440 |
| }, |
| { |
| "epoch": 7.807017543859649, |
| "grad_norm": 0.3265489637851715, |
| "learning_rate": 0.000765738196708228, |
| "loss": 0.0259, |
| "num_input_tokens_seen": 288416, |
| "step": 445 |
| }, |
| { |
| "epoch": 7.894736842105263, |
| "grad_norm": 0.21435128152370453, |
| "learning_rate": 0.0007592230023330069, |
| "loss": 0.0222, |
| "num_input_tokens_seen": 291968, |
| "step": 450 |
| }, |
| { |
| "epoch": 7.982456140350877, |
| "grad_norm": 0.022913899272680283, |
| "learning_rate": 0.000752647049032155, |
| "loss": 0.0113, |
| "num_input_tokens_seen": 295296, |
| "step": 455 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.15292111039161682, |
| "eval_runtime": 0.7464, |
| "eval_samples_per_second": 33.494, |
| "eval_steps_per_second": 9.378, |
| "num_input_tokens_seen": 295408, |
| "step": 456 |
| }, |
| { |
| "epoch": 8.070175438596491, |
| "grad_norm": 0.02062864601612091, |
| "learning_rate": 0.0007460118781343892, |
| "loss": 0.0045, |
| "num_input_tokens_seen": 298480, |
| "step": 460 |
| }, |
| { |
| "epoch": 8.157894736842104, |
| "grad_norm": 0.006704544648528099, |
| "learning_rate": 0.000739319044848358, |
| "loss": 0.078, |
| "num_input_tokens_seen": 301904, |
| "step": 465 |
| }, |
| { |
| "epoch": 8.24561403508772, |
| "grad_norm": 0.0441327840089798, |
| "learning_rate": 0.0007325701178981183, |
| "loss": 0.0203, |
| "num_input_tokens_seen": 305488, |
| "step": 470 |
| }, |
| { |
| "epoch": 8.333333333333334, |
| "grad_norm": 0.1119450256228447, |
| "learning_rate": 0.0007257666791554447, |
| "loss": 0.0179, |
| "num_input_tokens_seen": 308336, |
| "step": 475 |
| }, |
| { |
| "epoch": 8.421052631578947, |
| "grad_norm": 0.37555059790611267, |
| "learning_rate": 0.0007189103232690561, |
| "loss": 0.0517, |
| "num_input_tokens_seen": 311600, |
| "step": 480 |
| }, |
| { |
| "epoch": 8.508771929824562, |
| "grad_norm": 0.09913496673107147, |
| "learning_rate": 0.0007120026572908484, |
| "loss": 0.0194, |
| "num_input_tokens_seen": 315216, |
| "step": 485 |
| }, |
| { |
| "epoch": 8.596491228070175, |
| "grad_norm": 0.0433107353746891, |
| "learning_rate": 0.0007050453002992201, |
| "loss": 0.0256, |
| "num_input_tokens_seen": 318640, |
| "step": 490 |
| }, |
| { |
| "epoch": 8.68421052631579, |
| "grad_norm": 0.016323000192642212, |
| "learning_rate": 0.0006980398830195785, |
| "loss": 0.0051, |
| "num_input_tokens_seen": 321680, |
| "step": 495 |
| }, |
| { |
| "epoch": 8.771929824561404, |
| "grad_norm": 0.006579800043255091, |
| "learning_rate": 0.000690988047442116, |
| "loss": 0.0025, |
| "num_input_tokens_seen": 324464, |
| "step": 500 |
| }, |
| { |
| "epoch": 8.859649122807017, |
| "grad_norm": 0.00924243126064539, |
| "learning_rate": 0.0006838914464369467, |
| "loss": 0.0031, |
| "num_input_tokens_seen": 327888, |
| "step": 505 |
| }, |
| { |
| "epoch": 8.947368421052632, |
| "grad_norm": 0.007734250742942095, |
| "learning_rate": 0.0006767517433666918, |
| "loss": 0.0066, |
| "num_input_tokens_seen": 331312, |
| "step": 510 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.15188561379909515, |
| "eval_runtime": 0.7442, |
| "eval_samples_per_second": 33.592, |
| "eval_steps_per_second": 9.406, |
| "num_input_tokens_seen": 332648, |
| "step": 513 |
| }, |
| { |
| "epoch": 9.035087719298245, |
| "grad_norm": 0.005904178135097027, |
| "learning_rate": 0.0006695706116966074, |
| "loss": 0.0913, |
| "num_input_tokens_seen": 334120, |
| "step": 515 |
| }, |
| { |
| "epoch": 9.12280701754386, |
| "grad_norm": 0.0130391800776124, |
| "learning_rate": 0.0006623497346023419, |
| "loss": 0.0081, |
| "num_input_tokens_seen": 338280, |
| "step": 520 |
| }, |
| { |
| "epoch": 9.210526315789474, |
| "grad_norm": 0.030641527846455574, |
| "learning_rate": 0.0006550908045754194, |
| "loss": 0.0075, |
| "num_input_tokens_seen": 341064, |
| "step": 525 |
| }, |
| { |
| "epoch": 9.298245614035087, |
| "grad_norm": 0.018244441598653793, |
| "learning_rate": 0.0006477955230265393, |
| "loss": 0.0032, |
| "num_input_tokens_seen": 344552, |
| "step": 530 |
| }, |
| { |
| "epoch": 9.385964912280702, |
| "grad_norm": 0.006423134822398424, |
| "learning_rate": 0.0006404655998867848, |
| "loss": 0.0044, |
| "num_input_tokens_seen": 347592, |
| "step": 535 |
| }, |
| { |
| "epoch": 9.473684210526315, |
| "grad_norm": 0.0025237384252250195, |
| "learning_rate": 0.0006331027532068335, |
| "loss": 0.0032, |
| "num_input_tokens_seen": 350440, |
| "step": 540 |
| }, |
| { |
| "epoch": 9.56140350877193, |
| "grad_norm": 0.07135269790887833, |
| "learning_rate": 0.0006257087087542672, |
| "loss": 0.0025, |
| "num_input_tokens_seen": 353992, |
| "step": 545 |
| }, |
| { |
| "epoch": 9.649122807017545, |
| "grad_norm": 0.007862250320613384, |
| "learning_rate": 0.0006182851996090712, |
| "loss": 0.0442, |
| "num_input_tokens_seen": 357544, |
| "step": 550 |
| }, |
| { |
| "epoch": 9.736842105263158, |
| "grad_norm": 0.09623616188764572, |
| "learning_rate": 0.0006108339657574193, |
| "loss": 0.0439, |
| "num_input_tokens_seen": 360872, |
| "step": 555 |
| }, |
| { |
| "epoch": 9.824561403508772, |
| "grad_norm": 0.10430023074150085, |
| "learning_rate": 0.000603356753683842, |
| "loss": 0.0262, |
| "num_input_tokens_seen": 364200, |
| "step": 560 |
| }, |
| { |
| "epoch": 9.912280701754385, |
| "grad_norm": 0.0760573297739029, |
| "learning_rate": 0.0005958553159618693, |
| "loss": 0.0104, |
| "num_input_tokens_seen": 367656, |
| "step": 565 |
| }, |
| { |
| "epoch": 10.0, |
| "grad_norm": 0.046638406813144684, |
| "learning_rate": 0.0005883314108432481, |
| "loss": 0.0137, |
| "num_input_tokens_seen": 369976, |
| "step": 570 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.17922309041023254, |
| "eval_runtime": 0.7446, |
| "eval_samples_per_second": 33.575, |
| "eval_steps_per_second": 9.401, |
| "num_input_tokens_seen": 369976, |
| "step": 570 |
| }, |
| { |
| "epoch": 10.087719298245615, |
| "grad_norm": 0.042038898915052414, |
| "learning_rate": 0.0005807868018458274, |
| "loss": 0.0118, |
| "num_input_tokens_seen": 373304, |
| "step": 575 |
| }, |
| { |
| "epoch": 10.175438596491228, |
| "grad_norm": 0.5674363374710083, |
| "learning_rate": 0.0005732232573402109, |
| "loss": 0.0469, |
| "num_input_tokens_seen": 376632, |
| "step": 580 |
| }, |
| { |
| "epoch": 10.263157894736842, |
| "grad_norm": 0.04679185897111893, |
| "learning_rate": 0.0005656425501352691, |
| "loss": 0.0324, |
| "num_input_tokens_seen": 379608, |
| "step": 585 |
| }, |
| { |
| "epoch": 10.350877192982455, |
| "grad_norm": 0.044311270117759705, |
| "learning_rate": 0.0005580464570626152, |
| "loss": 0.0076, |
| "num_input_tokens_seen": 382296, |
| "step": 590 |
| }, |
| { |
| "epoch": 10.43859649122807, |
| "grad_norm": 0.010670201852917671, |
| "learning_rate": 0.0005504367585601342, |
| "loss": 0.0058, |
| "num_input_tokens_seen": 385880, |
| "step": 595 |
| }, |
| { |
| "epoch": 10.526315789473685, |
| "grad_norm": 0.22319240868091583, |
| "learning_rate": 0.0005428152382546695, |
| "loss": 0.018, |
| "num_input_tokens_seen": 388824, |
| "step": 600 |
| }, |
| { |
| "epoch": 10.614035087719298, |
| "grad_norm": 0.19307509064674377, |
| "learning_rate": 0.0005351836825439609, |
| "loss": 0.0099, |
| "num_input_tokens_seen": 391640, |
| "step": 605 |
| }, |
| { |
| "epoch": 10.701754385964913, |
| "grad_norm": 0.20500296354293823, |
| "learning_rate": 0.0005275438801779327, |
| "loss": 0.0048, |
| "num_input_tokens_seen": 394968, |
| "step": 610 |
| }, |
| { |
| "epoch": 10.789473684210526, |
| "grad_norm": 0.028505077585577965, |
| "learning_rate": 0.0005198976218394321, |
| "loss": 0.0016, |
| "num_input_tokens_seen": 398520, |
| "step": 615 |
| }, |
| { |
| "epoch": 10.87719298245614, |
| "grad_norm": 0.04338080435991287, |
| "learning_rate": 0.0005122466997245124, |
| "loss": 0.0015, |
| "num_input_tokens_seen": 402328, |
| "step": 620 |
| }, |
| { |
| "epoch": 10.964912280701755, |
| "grad_norm": 0.021871905773878098, |
| "learning_rate": 0.0005045929071223632, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 405880, |
| "step": 625 |
| }, |
| { |
| "epoch": 11.0, |
| "eval_loss": 0.2538582682609558, |
| "eval_runtime": 0.7471, |
| "eval_samples_per_second": 33.463, |
| "eval_steps_per_second": 9.37, |
| "num_input_tokens_seen": 406840, |
| "step": 627 |
| }, |
| { |
| "epoch": 11.052631578947368, |
| "grad_norm": 0.002679206430912018, |
| "learning_rate": 0.0004969380379949836, |
| "loss": 0.001, |
| "num_input_tokens_seen": 408632, |
| "step": 630 |
| }, |
| { |
| "epoch": 11.140350877192983, |
| "grad_norm": 0.6808245182037354, |
| "learning_rate": 0.0004892838865566986, |
| "loss": 0.0326, |
| "num_input_tokens_seen": 411544, |
| "step": 635 |
| }, |
| { |
| "epoch": 11.228070175438596, |
| "grad_norm": 0.010487693361938, |
| "learning_rate": 0.00048163224685361384, |
| "loss": 0.0014, |
| "num_input_tokens_seen": 414648, |
| "step": 640 |
| }, |
| { |
| "epoch": 11.31578947368421, |
| "grad_norm": 0.01465911790728569, |
| "learning_rate": 0.0004739849123431138, |
| "loss": 0.0017, |
| "num_input_tokens_seen": 418968, |
| "step": 645 |
| }, |
| { |
| "epoch": 11.403508771929825, |
| "grad_norm": 0.015837009996175766, |
| "learning_rate": 0.00046634367547349433, |
| "loss": 0.0011, |
| "num_input_tokens_seen": 422584, |
| "step": 650 |
| }, |
| { |
| "epoch": 11.491228070175438, |
| "grad_norm": 0.02893809974193573, |
| "learning_rate": 0.0004587103272638339, |
| "loss": 0.0035, |
| "num_input_tokens_seen": 426136, |
| "step": 655 |
| }, |
| { |
| "epoch": 11.578947368421053, |
| "grad_norm": 0.0007099352078512311, |
| "learning_rate": 0.0004510866568841981, |
| "loss": 0.0047, |
| "num_input_tokens_seen": 429752, |
| "step": 660 |
| }, |
| { |
| "epoch": 11.666666666666666, |
| "grad_norm": 0.10449596494436264, |
| "learning_rate": 0.0004434744512362797, |
| "loss": 0.0034, |
| "num_input_tokens_seen": 432920, |
| "step": 665 |
| }, |
| { |
| "epoch": 11.75438596491228, |
| "grad_norm": 0.03190397843718529, |
| "learning_rate": 0.00043587549453456836, |
| "loss": 0.0018, |
| "num_input_tokens_seen": 435608, |
| "step": 670 |
| }, |
| { |
| "epoch": 11.842105263157894, |
| "grad_norm": 0.0037696484941989183, |
| "learning_rate": 0.00042829156788815195, |
| "loss": 0.006, |
| "num_input_tokens_seen": 438744, |
| "step": 675 |
| }, |
| { |
| "epoch": 11.929824561403509, |
| "grad_norm": 0.005847765598446131, |
| "learning_rate": 0.0004207244488832429, |
| "loss": 0.0023, |
| "num_input_tokens_seen": 442200, |
| "step": 680 |
| }, |
| { |
| "epoch": 12.0, |
| "eval_loss": 0.2909103035926819, |
| "eval_runtime": 0.7451, |
| "eval_samples_per_second": 33.553, |
| "eval_steps_per_second": 9.395, |
| "num_input_tokens_seen": 444728, |
| "step": 684 |
| }, |
| { |
| "epoch": 12.017543859649123, |
| "grad_norm": 0.014109612442553043, |
| "learning_rate": 0.00041317591116653486, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 445624, |
| "step": 685 |
| }, |
| { |
| "epoch": 12.105263157894736, |
| "grad_norm": 0.0011010151356458664, |
| "learning_rate": 0.00040564772402947784, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 448632, |
| "step": 690 |
| }, |
| { |
| "epoch": 12.192982456140351, |
| "grad_norm": 0.000885249231941998, |
| "learning_rate": 0.00039814165199357807, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 452248, |
| "step": 695 |
| }, |
| { |
| "epoch": 12.280701754385966, |
| "grad_norm": 0.38688522577285767, |
| "learning_rate": 0.00039065945439681213, |
| "loss": 0.0053, |
| "num_input_tokens_seen": 455544, |
| "step": 700 |
| }, |
| { |
| "epoch": 12.368421052631579, |
| "grad_norm": 0.0028335454408079386, |
| "learning_rate": 0.0003832028849812607, |
| "loss": 0.0007, |
| "num_input_tokens_seen": 458520, |
| "step": 705 |
| }, |
| { |
| "epoch": 12.456140350877194, |
| "grad_norm": 0.0012163643259555101, |
| "learning_rate": 0.00037577369148204934, |
| "loss": 0.0021, |
| "num_input_tokens_seen": 462264, |
| "step": 710 |
| }, |
| { |
| "epoch": 12.543859649122806, |
| "grad_norm": 0.001764231943525374, |
| "learning_rate": 0.00036837361521770053, |
| "loss": 0.0009, |
| "num_input_tokens_seen": 465624, |
| "step": 715 |
| }, |
| { |
| "epoch": 12.631578947368421, |
| "grad_norm": 0.018389353528618813, |
| "learning_rate": 0.00036100439068198676, |
| "loss": 0.0016, |
| "num_input_tokens_seen": 468536, |
| "step": 720 |
| }, |
| { |
| "epoch": 12.719298245614034, |
| "grad_norm": 0.014804023317992687, |
| "learning_rate": 0.00035366774513738707, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 472408, |
| "step": 725 |
| }, |
| { |
| "epoch": 12.807017543859649, |
| "grad_norm": 0.0032388227991759777, |
| "learning_rate": 0.0003463653982102347, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 475416, |
| "step": 730 |
| }, |
| { |
| "epoch": 12.894736842105264, |
| "grad_norm": 0.0006264116382226348, |
| "learning_rate": 0.00033909906148765724, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 478712, |
| "step": 735 |
| }, |
| { |
| "epoch": 12.982456140350877, |
| "grad_norm": 0.013625812716782093, |
| "learning_rate": 0.00033187043811639863, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 481592, |
| "step": 740 |
| }, |
| { |
| "epoch": 13.0, |
| "eval_loss": 0.3705059587955475, |
| "eval_runtime": 0.7446, |
| "eval_samples_per_second": 33.577, |
| "eval_steps_per_second": 9.401, |
| "num_input_tokens_seen": 481720, |
| "step": 741 |
| }, |
| { |
| "epoch": 13.070175438596491, |
| "grad_norm": 0.0008147696498781443, |
| "learning_rate": 0.00032468122240362287, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 484856, |
| "step": 745 |
| }, |
| { |
| "epoch": 13.157894736842104, |
| "grad_norm": 0.005378706846386194, |
| "learning_rate": 0.00031753309941978615, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 487480, |
| "step": 750 |
| }, |
| { |
| "epoch": 13.24561403508772, |
| "grad_norm": 0.002744952216744423, |
| "learning_rate": 0.0003104277446036764, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 490520, |
| "step": 755 |
| }, |
| { |
| "epoch": 13.333333333333334, |
| "grad_norm": 0.0012103226035833359, |
| "learning_rate": 0.00030336682336970847, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 493976, |
| "step": 760 |
| }, |
| { |
| "epoch": 13.421052631578947, |
| "grad_norm": 0.0006653820746578276, |
| "learning_rate": 0.0002963519907175713, |
| "loss": 0.0005, |
| "num_input_tokens_seen": 497080, |
| "step": 765 |
| }, |
| { |
| "epoch": 13.508771929824562, |
| "grad_norm": 0.0035558794625103474, |
| "learning_rate": 0.00028938489084431363, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 500792, |
| "step": 770 |
| }, |
| { |
| "epoch": 13.596491228070175, |
| "grad_norm": 0.013601898215711117, |
| "learning_rate": 0.0002824671567589635, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 503832, |
| "step": 775 |
| }, |
| { |
| "epoch": 13.68421052631579, |
| "grad_norm": 0.0014382227091118693, |
| "learning_rate": 0.00027560040989976894, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 507256, |
| "step": 780 |
| }, |
| { |
| "epoch": 13.771929824561404, |
| "grad_norm": 0.0005540886195376515, |
| "learning_rate": 0.0002687862597541523, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 510616, |
| "step": 785 |
| }, |
| { |
| "epoch": 13.859649122807017, |
| "grad_norm": 0.0010458007454872131, |
| "learning_rate": 0.0002620263034814632, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 514200, |
| "step": 790 |
| }, |
| { |
| "epoch": 13.947368421052632, |
| "grad_norm": 0.06604114919900894, |
| "learning_rate": 0.00025532212553862446, |
| "loss": 0.0008, |
| "num_input_tokens_seen": 517400, |
| "step": 795 |
| }, |
| { |
| "epoch": 14.0, |
| "eval_loss": 0.36595970392227173, |
| "eval_runtime": 0.746, |
| "eval_samples_per_second": 33.512, |
| "eval_steps_per_second": 9.383, |
| "num_input_tokens_seen": 518664, |
| "step": 798 |
| }, |
| { |
| "epoch": 14.035087719298245, |
| "grad_norm": 0.0014471168396994472, |
| "learning_rate": 0.000248675297308751, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 519848, |
| "step": 800 |
| }, |
| { |
| "epoch": 14.12280701754386, |
| "grad_norm": 0.002391919493675232, |
| "learning_rate": 0.00024208737673283814, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 523176, |
| "step": 805 |
| }, |
| { |
| "epoch": 14.210526315789474, |
| "grad_norm": 0.007587758358567953, |
| "learning_rate": 0.00023555990794459542, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 526280, |
| "step": 810 |
| }, |
| { |
| "epoch": 14.298245614035087, |
| "grad_norm": 0.0007305729086510837, |
| "learning_rate": 0.00022909442090852144, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 529160, |
| "step": 815 |
| }, |
| { |
| "epoch": 14.385964912280702, |
| "grad_norm": 0.005135450512170792, |
| "learning_rate": 0.0002226924310612956, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 533256, |
| "step": 820 |
| }, |
| { |
| "epoch": 14.473684210526315, |
| "grad_norm": 0.0027909493073821068, |
| "learning_rate": 0.00021635543895657866, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 536712, |
| "step": 825 |
| }, |
| { |
| "epoch": 14.56140350877193, |
| "grad_norm": 0.0009187912801280618, |
| "learning_rate": 0.00021008492991329863, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 539912, |
| "step": 830 |
| }, |
| { |
| "epoch": 14.649122807017545, |
| "grad_norm": 0.0006335495854727924, |
| "learning_rate": 0.00020388237366751006, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 543336, |
| "step": 835 |
| }, |
| { |
| "epoch": 14.736842105263158, |
| "grad_norm": 0.003865489037707448, |
| "learning_rate": 0.0001977492240279035, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 547496, |
| "step": 840 |
| }, |
| { |
| "epoch": 14.824561403508772, |
| "grad_norm": 0.0015492694219574332, |
| "learning_rate": 0.0001916869185350505, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 550280, |
| "step": 845 |
| }, |
| { |
| "epoch": 14.912280701754385, |
| "grad_norm": 0.0075766416266560555, |
| "learning_rate": 0.00018569687812445895, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 553672, |
| "step": 850 |
| }, |
| { |
| "epoch": 15.0, |
| "grad_norm": 0.0077131628058850765, |
| "learning_rate": 0.00017978050679352359, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 555728, |
| "step": 855 |
| }, |
| { |
| "epoch": 15.0, |
| "eval_loss": 0.36777299642562866, |
| "eval_runtime": 0.7431, |
| "eval_samples_per_second": 33.641, |
| "eval_steps_per_second": 9.419, |
| "num_input_tokens_seen": 555728, |
| "step": 855 |
| }, |
| { |
| "epoch": 15.087719298245615, |
| "grad_norm": 0.0028925647493451834, |
| "learning_rate": 0.00017393919127244346, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 559312, |
| "step": 860 |
| }, |
| { |
| "epoch": 15.175438596491228, |
| "grad_norm": 0.000598029000684619, |
| "learning_rate": 0.00016817430069918936, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 563056, |
| "step": 865 |
| }, |
| { |
| "epoch": 15.263157894736842, |
| "grad_norm": 0.0024406970478594303, |
| "learning_rate": 0.00016248718629859244, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 566224, |
| "step": 870 |
| }, |
| { |
| "epoch": 15.350877192982455, |
| "grad_norm": 0.0006085984059609473, |
| "learning_rate": 0.00015687918106563326, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 569776, |
| "step": 875 |
| }, |
| { |
| "epoch": 15.43859649122807, |
| "grad_norm": 0.0007883381913416088, |
| "learning_rate": 0.0001513515994530023, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 573360, |
| "step": 880 |
| }, |
| { |
| "epoch": 15.526315789473685, |
| "grad_norm": 0.002710202243179083, |
| "learning_rate": 0.00014590573706300782, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 576144, |
| "step": 885 |
| }, |
| { |
| "epoch": 15.614035087719298, |
| "grad_norm": 0.010271117091178894, |
| "learning_rate": 0.00014054287034390045, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 579088, |
| "step": 890 |
| }, |
| { |
| "epoch": 15.701754385964913, |
| "grad_norm": 0.003705398179590702, |
| "learning_rate": 0.00013526425629068966, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 582032, |
| "step": 895 |
| }, |
| { |
| "epoch": 15.789473684210526, |
| "grad_norm": 0.0013794214464724064, |
| "learning_rate": 0.00013007113215051673, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 585584, |
| "step": 900 |
| }, |
| { |
| "epoch": 15.87719298245614, |
| "grad_norm": 0.00568090844899416, |
| "learning_rate": 0.00012496471513265967, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 588944, |
| "step": 905 |
| }, |
| { |
| "epoch": 15.964912280701755, |
| "grad_norm": 0.001053197542205453, |
| "learning_rate": 0.00011994620212323176, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 592112, |
| "step": 910 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_loss": 0.3890913426876068, |
| "eval_runtime": 0.7441, |
| "eval_samples_per_second": 33.596, |
| "eval_steps_per_second": 9.407, |
| "num_input_tokens_seen": 593096, |
| "step": 912 |
| }, |
| { |
| "epoch": 16.05263157894737, |
| "grad_norm": 0.001349198049865663, |
| "learning_rate": 0.00011501676940464645, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 594440, |
| "step": 915 |
| }, |
| { |
| "epoch": 16.140350877192983, |
| "grad_norm": 0.0005845442065037787, |
| "learning_rate": 0.00011017757237990877, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 598440, |
| "step": 920 |
| }, |
| { |
| "epoch": 16.228070175438596, |
| "grad_norm": 0.007813757285475731, |
| "learning_rate": 0.00010542974530180327, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 602152, |
| "step": 925 |
| }, |
| { |
| "epoch": 16.31578947368421, |
| "grad_norm": 0.001801942940801382, |
| "learning_rate": 0.00010077440100703683, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 605736, |
| "step": 930 |
| }, |
| { |
| "epoch": 16.403508771929825, |
| "grad_norm": 0.00401125755161047, |
| "learning_rate": 9.621263065540364e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 608520, |
| "step": 935 |
| }, |
| { |
| "epoch": 16.49122807017544, |
| "grad_norm": 0.0013953890884295106, |
| "learning_rate": 9.174550347402855e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 612040, |
| "step": 940 |
| }, |
| { |
| "epoch": 16.57894736842105, |
| "grad_norm": 0.0009140498586930335, |
| "learning_rate": 8.737406650675333e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 614920, |
| "step": 945 |
| }, |
| { |
| "epoch": 16.666666666666668, |
| "grad_norm": 0.0005205461056903005, |
| "learning_rate": 8.309934436872074e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 618440, |
| "step": 950 |
| }, |
| { |
| "epoch": 16.75438596491228, |
| "grad_norm": 0.0019901059567928314, |
| "learning_rate": 7.89223390062172e-05, |
| "loss": 0.0006, |
| "num_input_tokens_seen": 621320, |
| "step": 955 |
| }, |
| { |
| "epoch": 16.842105263157894, |
| "grad_norm": 0.0006933985278010368, |
| "learning_rate": 7.4844029461827e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 624648, |
| "step": 960 |
| }, |
| { |
| "epoch": 16.92982456140351, |
| "grad_norm": 0.006941209081560373, |
| "learning_rate": 7.086537164495688e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 628232, |
| "step": 965 |
| }, |
| { |
| "epoch": 17.0, |
| "eval_loss": 0.37933316826820374, |
| "eval_runtime": 0.7447, |
| "eval_samples_per_second": 33.571, |
| "eval_steps_per_second": 9.4, |
| "num_input_tokens_seen": 629760, |
| "step": 969 |
| }, |
| { |
| "epoch": 17.017543859649123, |
| "grad_norm": 0.0028358723502606153, |
| "learning_rate": 6.698729810778065e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 630528, |
| "step": 970 |
| }, |
| { |
| "epoch": 17.105263157894736, |
| "grad_norm": 0.0003263541730120778, |
| "learning_rate": 6.321071782666077e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 633600, |
| "step": 975 |
| }, |
| { |
| "epoch": 17.19298245614035, |
| "grad_norm": 0.0036093771923333406, |
| "learning_rate": 5.953651598909332e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 637024, |
| "step": 980 |
| }, |
| { |
| "epoch": 17.280701754385966, |
| "grad_norm": 0.002093312796205282, |
| "learning_rate": 5.596555378623125e-05, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 640160, |
| "step": 985 |
| }, |
| { |
| "epoch": 17.36842105263158, |
| "grad_norm": 0.002227015094831586, |
| "learning_rate": 5.2498668211030166e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 643040, |
| "step": 990 |
| }, |
| { |
| "epoch": 17.45614035087719, |
| "grad_norm": 0.001337834750302136, |
| "learning_rate": 4.913667186206722e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 646336, |
| "step": 995 |
| }, |
| { |
| "epoch": 17.54385964912281, |
| "grad_norm": 0.0036892376374453306, |
| "learning_rate": 4.588035275307689e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 649984, |
| "step": 1000 |
| }, |
| { |
| "epoch": 17.63157894736842, |
| "grad_norm": 0.0012848501792177558, |
| "learning_rate": 4.273047412824954e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 653632, |
| "step": 1005 |
| }, |
| { |
| "epoch": 17.719298245614034, |
| "grad_norm": 0.0038990762550383806, |
| "learning_rate": 3.9687774283335975e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 656960, |
| "step": 1010 |
| }, |
| { |
| "epoch": 17.80701754385965, |
| "grad_norm": 0.00040850613731890917, |
| "learning_rate": 3.675296639259912e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 659840, |
| "step": 1015 |
| }, |
| { |
| "epoch": 17.894736842105264, |
| "grad_norm": 0.01073791366070509, |
| "learning_rate": 3.392673834165388e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 663872, |
| "step": 1020 |
| }, |
| { |
| "epoch": 17.982456140350877, |
| "grad_norm": 0.004404230508953333, |
| "learning_rate": 3.120975256623465e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 667264, |
| "step": 1025 |
| }, |
| { |
| "epoch": 18.0, |
| "eval_loss": 0.38870540261268616, |
| "eval_runtime": 0.7479, |
| "eval_samples_per_second": 33.429, |
| "eval_steps_per_second": 9.36, |
| "num_input_tokens_seen": 667432, |
| "step": 1026 |
| }, |
| { |
| "epoch": 18.07017543859649, |
| "grad_norm": 0.0011177717242389917, |
| "learning_rate": 2.8602645896928293e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 669544, |
| "step": 1030 |
| }, |
| { |
| "epoch": 18.157894736842106, |
| "grad_norm": 0.0034124741796404123, |
| "learning_rate": 2.610602940990797e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 672776, |
| "step": 1035 |
| }, |
| { |
| "epoch": 18.24561403508772, |
| "grad_norm": 0.004636778496205807, |
| "learning_rate": 2.3720488283703547e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 676008, |
| "step": 1040 |
| }, |
| { |
| "epoch": 18.333333333333332, |
| "grad_norm": 0.002139877527952194, |
| "learning_rate": 2.144658166204294e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 679848, |
| "step": 1045 |
| }, |
| { |
| "epoch": 18.42105263157895, |
| "grad_norm": 0.0005989308701828122, |
| "learning_rate": 1.9284842522794942e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 683144, |
| "step": 1050 |
| }, |
| { |
| "epoch": 18.50877192982456, |
| "grad_norm": 0.0007390376995317638, |
| "learning_rate": 1.7235777553045283e-05, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 686280, |
| "step": 1055 |
| }, |
| { |
| "epoch": 18.596491228070175, |
| "grad_norm": 0.0010978280333802104, |
| "learning_rate": 1.5299867030334813e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 689480, |
| "step": 1060 |
| }, |
| { |
| "epoch": 18.68421052631579, |
| "grad_norm": 0.0012614470906555653, |
| "learning_rate": 1.3477564710088097e-05, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 692296, |
| "step": 1065 |
| }, |
| { |
| "epoch": 18.771929824561404, |
| "grad_norm": 0.0010993927717208862, |
| "learning_rate": 1.1769297719258221e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 695912, |
| "step": 1070 |
| }, |
| { |
| "epoch": 18.859649122807017, |
| "grad_norm": 0.0016077933833003044, |
| "learning_rate": 1.0175466456213034e-05, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 699432, |
| "step": 1075 |
| }, |
| { |
| "epoch": 18.94736842105263, |
| "grad_norm": 0.013521168380975723, |
| "learning_rate": 8.696444496886502e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 703080, |
| "step": 1080 |
| }, |
| { |
| "epoch": 19.0, |
| "eval_loss": 0.39144256711006165, |
| "eval_runtime": 0.7477, |
| "eval_samples_per_second": 33.437, |
| "eval_steps_per_second": 9.362, |
| "num_input_tokens_seen": 704816, |
| "step": 1083 |
| }, |
| { |
| "epoch": 19.035087719298247, |
| "grad_norm": 0.004061270505189896, |
| "learning_rate": 7.332578507216469e-06, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 706192, |
| "step": 1085 |
| }, |
| { |
| "epoch": 19.12280701754386, |
| "grad_norm": 0.005511495750397444, |
| "learning_rate": 6.084188161890325e-06, |
| "loss": 0.0004, |
| "num_input_tokens_seen": 709072, |
| "step": 1090 |
| }, |
| { |
| "epoch": 19.210526315789473, |
| "grad_norm": 0.0010893200524151325, |
| "learning_rate": 4.95156606941688e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 712496, |
| "step": 1095 |
| }, |
| { |
| "epoch": 19.29824561403509, |
| "grad_norm": 0.0010816626017913222, |
| "learning_rate": 3.9349777035421194e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 715056, |
| "step": 1100 |
| }, |
| { |
| "epoch": 19.385964912280702, |
| "grad_norm": 0.00045030307956039906, |
| "learning_rate": 3.034661341025258e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 718416, |
| "step": 1105 |
| }, |
| { |
| "epoch": 19.473684210526315, |
| "grad_norm": 0.0028827146161347628, |
| "learning_rate": 2.250828005789518e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 721744, |
| "step": 1110 |
| }, |
| { |
| "epoch": 19.56140350877193, |
| "grad_norm": 0.001782647450454533, |
| "learning_rate": 1.5836614194602028e-06, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 725328, |
| "step": 1115 |
| }, |
| { |
| "epoch": 19.649122807017545, |
| "grad_norm": 0.0003462164895609021, |
| "learning_rate": 1.033317958302693e-06, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 728976, |
| "step": 1120 |
| }, |
| { |
| "epoch": 19.736842105263158, |
| "grad_norm": 0.010955123230814934, |
| "learning_rate": 5.999266165694906e-07, |
| "loss": 0.0003, |
| "num_input_tokens_seen": 732368, |
| "step": 1125 |
| }, |
| { |
| "epoch": 19.82456140350877, |
| "grad_norm": 0.001071834471076727, |
| "learning_rate": 2.8358897626556966e-07, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 735440, |
| "step": 1130 |
| }, |
| { |
| "epoch": 19.912280701754387, |
| "grad_norm": 0.0024311819579452276, |
| "learning_rate": 8.437918333864537e-08, |
| "loss": 0.0002, |
| "num_input_tokens_seen": 739120, |
| "step": 1135 |
| }, |
| { |
| "epoch": 20.0, |
| "grad_norm": 0.0004392106784507632, |
| "learning_rate": 2.343930299963937e-09, |
| "loss": 0.0001, |
| "num_input_tokens_seen": 742296, |
| "step": 1140 |
| }, |
| { |
| "epoch": 20.0, |
| "eval_loss": 0.39334043860435486, |
| "eval_runtime": 0.7462, |
| "eval_samples_per_second": 33.501, |
| "eval_steps_per_second": 9.38, |
| "num_input_tokens_seen": 742296, |
| "step": 1140 |
| }, |
| { |
| "epoch": 20.0, |
| "num_input_tokens_seen": 742296, |
| "step": 1140, |
| "total_flos": 3.342525338596147e+16, |
| "train_loss": 0.11720111153829892, |
| "train_runtime": 292.3442, |
| "train_samples_per_second": 15.393, |
| "train_steps_per_second": 3.9 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 1140, |
| "num_input_tokens_seen": 742296, |
| "num_train_epochs": 20, |
| "save_steps": 57, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.342525338596147e+16, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
