| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.9914040114613183, | |
| "eval_steps": 50, | |
| "global_step": 261, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05730659025787966, | |
| "grad_norm": 13.627805370893373, | |
| "learning_rate": 5e-07, | |
| "loss": 1.2695, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.11461318051575932, | |
| "grad_norm": 8.516113291182938, | |
| "learning_rate": 1e-06, | |
| "loss": 1.1722, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17191977077363896, | |
| "grad_norm": 5.228917291507613, | |
| "learning_rate": 9.990212076323586e-07, | |
| "loss": 1.0671, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.22922636103151864, | |
| "grad_norm": 3.994531672788626, | |
| "learning_rate": 9.9608866266743e-07, | |
| "loss": 1.0138, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.28653295128939826, | |
| "grad_norm": 3.8970974428882528, | |
| "learning_rate": 9.912138465157323e-07, | |
| "loss": 1.0128, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3438395415472779, | |
| "grad_norm": 3.5838024972775804, | |
| "learning_rate": 9.84415844908637e-07, | |
| "loss": 0.9834, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.40114613180515757, | |
| "grad_norm": 3.5452348908066513, | |
| "learning_rate": 9.757212731744973e-07, | |
| "loss": 0.9669, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.4584527220630373, | |
| "grad_norm": 3.5843368374533697, | |
| "learning_rate": 9.65164172035126e-07, | |
| "loss": 0.9693, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5157593123209169, | |
| "grad_norm": 3.428042968685717, | |
| "learning_rate": 9.527858743306018e-07, | |
| "loss": 0.9532, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5730659025787965, | |
| "grad_norm": 3.4859824244397886, | |
| "learning_rate": 9.386348431941952e-07, | |
| "loss": 0.9473, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5730659025787965, | |
| "eval_loss": 0.9285200834274292, | |
| "eval_runtime": 42.6562, | |
| "eval_samples_per_second": 58.116, | |
| "eval_steps_per_second": 0.914, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6303724928366762, | |
| "grad_norm": 3.5547459569878144, | |
| "learning_rate": 9.227664823109882e-07, | |
| "loss": 0.9391, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.6876790830945558, | |
| "grad_norm": 3.436701336864977, | |
| "learning_rate": 9.052429190030588e-07, | |
| "loss": 0.9422, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7449856733524355, | |
| "grad_norm": 3.503857788745808, | |
| "learning_rate": 8.861327609904857e-07, | |
| "loss": 0.943, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.8022922636103151, | |
| "grad_norm": 3.58484631574656, | |
| "learning_rate": 8.655108277804975e-07, | |
| "loss": 0.9253, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8595988538681948, | |
| "grad_norm": 3.4761668389890055, | |
| "learning_rate": 8.434578577364217e-07, | |
| "loss": 0.9285, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9169054441260746, | |
| "grad_norm": 3.9387084351724977, | |
| "learning_rate": 8.200601919733105e-07, | |
| "loss": 0.8996, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9742120343839542, | |
| "grad_norm": 3.5121314302877518, | |
| "learning_rate": 7.954094363178421e-07, | |
| "loss": 0.895, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.0315186246418337, | |
| "grad_norm": 4.367748237913567, | |
| "learning_rate": 7.696021026559849e-07, | |
| "loss": 0.8506, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0888252148997135, | |
| "grad_norm": 4.08819656393782, | |
| "learning_rate": 7.427392310726087e-07, | |
| "loss": 0.7965, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.146131805157593, | |
| "grad_norm": 3.9211880538603254, | |
| "learning_rate": 7.149259942624286e-07, | |
| "loss": 0.7919, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.146131805157593, | |
| "eval_loss": 0.887869119644165, | |
| "eval_runtime": 42.996, | |
| "eval_samples_per_second": 57.657, | |
| "eval_steps_per_second": 0.907, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2034383954154728, | |
| "grad_norm": 3.8896671321207696, | |
| "learning_rate": 6.862712857610811e-07, | |
| "loss": 0.7851, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.2607449856733524, | |
| "grad_norm": 3.972314483632093, | |
| "learning_rate": 6.568872936084788e-07, | |
| "loss": 0.7942, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.3180515759312321, | |
| "grad_norm": 3.712979983054785, | |
| "learning_rate": 6.26889061113621e-07, | |
| "loss": 0.7818, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.3753581661891117, | |
| "grad_norm": 4.088396831337597, | |
| "learning_rate": 5.963940364405425e-07, | |
| "loss": 0.7888, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.4326647564469914, | |
| "grad_norm": 3.664011338983596, | |
| "learning_rate": 5.655216127788472e-07, | |
| "loss": 0.7783, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.4899713467048712, | |
| "grad_norm": 3.944962672437967, | |
| "learning_rate": 5.343926608991379e-07, | |
| "loss": 0.7718, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5472779369627507, | |
| "grad_norm": 3.9718807100230364, | |
| "learning_rate": 5.031290559234649e-07, | |
| "loss": 0.7789, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.6045845272206303, | |
| "grad_norm": 3.964006687061035, | |
| "learning_rate": 4.718532001635686e-07, | |
| "loss": 0.7749, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.66189111747851, | |
| "grad_norm": 5.307390030470514, | |
| "learning_rate": 4.406875438950861e-07, | |
| "loss": 0.7639, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.7191977077363898, | |
| "grad_norm": 3.8075901896947526, | |
| "learning_rate": 4.097541059439698e-07, | |
| "loss": 0.7632, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.7191977077363898, | |
| "eval_loss": 0.8642405867576599, | |
| "eval_runtime": 42.2443, | |
| "eval_samples_per_second": 58.682, | |
| "eval_steps_per_second": 0.923, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.7765042979942693, | |
| "grad_norm": 3.9277854050418446, | |
| "learning_rate": 3.7917399596210535e-07, | |
| "loss": 0.7688, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.8338108882521489, | |
| "grad_norm": 3.805908491229589, | |
| "learning_rate": 3.490669402625007e-07, | |
| "loss": 0.7653, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.8911174785100286, | |
| "grad_norm": 3.9581522213139366, | |
| "learning_rate": 3.195508130704795e-07, | |
| "loss": 0.7608, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.9484240687679084, | |
| "grad_norm": 4.004608121816857, | |
| "learning_rate": 2.9074117502611296e-07, | |
| "loss": 0.756, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.005730659025788, | |
| "grad_norm": 4.507845402680472, | |
| "learning_rate": 2.6275082074473075e-07, | |
| "loss": 0.7396, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.0630372492836675, | |
| "grad_norm": 4.307602747932898, | |
| "learning_rate": 2.3568933720688543e-07, | |
| "loss": 0.6778, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.1203438395415475, | |
| "grad_norm": 4.299030469560283, | |
| "learning_rate": 2.096626747067527e-07, | |
| "loss": 0.6679, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.177650429799427, | |
| "grad_norm": 4.282112579078971, | |
| "learning_rate": 1.8477273203877398e-07, | |
| "loss": 0.6878, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.2349570200573066, | |
| "grad_norm": 4.205893539800291, | |
| "learning_rate": 1.6111695754660664e-07, | |
| "loss": 0.6842, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.292263610315186, | |
| "grad_norm": 4.098098081505987, | |
| "learning_rate": 1.3878796759634542e-07, | |
| "loss": 0.685, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.292263610315186, | |
| "eval_loss": 0.8616007566452026, | |
| "eval_runtime": 42.3579, | |
| "eval_samples_per_second": 58.525, | |
| "eval_steps_per_second": 0.921, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.349570200573066, | |
| "grad_norm": 4.186543339384142, | |
| "learning_rate": 1.1787318396775186e-07, | |
| "loss": 0.6859, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.4068767908309456, | |
| "grad_norm": 4.433832809489656, | |
| "learning_rate": 9.845449158317215e-08, | |
| "loss": 0.6921, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.464183381088825, | |
| "grad_norm": 4.134230624778858, | |
| "learning_rate": 8.060791791418886e-08, | |
| "loss": 0.6699, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.5214899713467047, | |
| "grad_norm": 4.180670228448671, | |
| "learning_rate": 6.440333532118503e-08, | |
| "loss": 0.6842, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.5787965616045847, | |
| "grad_norm": 4.293320947026629, | |
| "learning_rate": 4.990418749121178e-08, | |
| "loss": 0.6743, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.6361031518624642, | |
| "grad_norm": 4.311294604114836, | |
| "learning_rate": 3.716724104520247e-08, | |
| "loss": 0.6757, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.693409742120344, | |
| "grad_norm": 4.080010522643778, | |
| "learning_rate": 2.624236328703061e-08, | |
| "loss": 0.6957, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 2.7507163323782233, | |
| "grad_norm": 3.9794059196967106, | |
| "learning_rate": 1.7172326964564775e-08, | |
| "loss": 0.672, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.8080229226361033, | |
| "grad_norm": 4.222641685902728, | |
| "learning_rate": 9.992642807111484e-09, | |
| "loss": 0.6896, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 2.865329512893983, | |
| "grad_norm": 4.2167318838478565, | |
| "learning_rate": 4.7314204948923354e-09, | |
| "loss": 0.6852, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.865329512893983, | |
| "eval_loss": 0.8607162237167358, | |
| "eval_runtime": 42.3035, | |
| "eval_samples_per_second": 58.6, | |
| "eval_steps_per_second": 0.922, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.9226361031518624, | |
| "grad_norm": 4.130630949457763, | |
| "learning_rate": 1.4092586048820575e-09, | |
| "loss": 0.6781, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 2.9799426934097424, | |
| "grad_norm": 4.129172689259573, | |
| "learning_rate": 3.91639638886998e-11, | |
| "loss": 0.6878, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.9914040114613183, | |
| "step": 261, | |
| "total_flos": 1538740350615552.0, | |
| "train_loss": 0.8157192728985315, | |
| "train_runtime": 3557.8908, | |
| "train_samples_per_second": 18.813, | |
| "train_steps_per_second": 0.073 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 261, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1538740350615552.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |