| { | |
| "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": 13.615523112232149, | |
| "learning_rate": 5e-07, | |
| "loss": 1.2695, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.11461318051575932, | |
| "grad_norm": 8.507104558172857, | |
| "learning_rate": 1e-06, | |
| "loss": 1.1722, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17191977077363896, | |
| "grad_norm": 5.216528499988659, | |
| "learning_rate": 9.990212076323586e-07, | |
| "loss": 1.067, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.22922636103151864, | |
| "grad_norm": 3.9965836016587044, | |
| "learning_rate": 9.9608866266743e-07, | |
| "loss": 1.0137, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.28653295128939826, | |
| "grad_norm": 3.90312687677931, | |
| "learning_rate": 9.912138465157323e-07, | |
| "loss": 1.0128, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.3438395415472779, | |
| "grad_norm": 3.5852669105149593, | |
| "learning_rate": 9.84415844908637e-07, | |
| "loss": 0.9835, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.40114613180515757, | |
| "grad_norm": 3.544317847930178, | |
| "learning_rate": 9.757212731744973e-07, | |
| "loss": 0.967, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.4584527220630373, | |
| "grad_norm": 3.581414511816451, | |
| "learning_rate": 9.65164172035126e-07, | |
| "loss": 0.9694, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5157593123209169, | |
| "grad_norm": 3.4337266155445634, | |
| "learning_rate": 9.527858743306018e-07, | |
| "loss": 0.9532, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.5730659025787965, | |
| "grad_norm": 3.4769930108810807, | |
| "learning_rate": 9.386348431941952e-07, | |
| "loss": 0.9472, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5730659025787965, | |
| "eval_loss": 0.9284669160842896, | |
| "eval_runtime": 42.918, | |
| "eval_samples_per_second": 57.761, | |
| "eval_steps_per_second": 0.909, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6303724928366762, | |
| "grad_norm": 3.550987444482616, | |
| "learning_rate": 9.227664823109882e-07, | |
| "loss": 0.9391, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.6876790830945558, | |
| "grad_norm": 3.4297326670171917, | |
| "learning_rate": 9.052429190030588e-07, | |
| "loss": 0.9422, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7449856733524355, | |
| "grad_norm": 3.507501066789746, | |
| "learning_rate": 8.861327609904857e-07, | |
| "loss": 0.9428, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.8022922636103151, | |
| "grad_norm": 3.5906687662141117, | |
| "learning_rate": 8.655108277804975e-07, | |
| "loss": 0.9253, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8595988538681948, | |
| "grad_norm": 3.500303702885294, | |
| "learning_rate": 8.434578577364217e-07, | |
| "loss": 0.9285, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.9169054441260746, | |
| "grad_norm": 3.9526314008449996, | |
| "learning_rate": 8.200601919733105e-07, | |
| "loss": 0.8996, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9742120343839542, | |
| "grad_norm": 3.516323615141867, | |
| "learning_rate": 7.954094363178421e-07, | |
| "loss": 0.895, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.0315186246418337, | |
| "grad_norm": 4.35173055618085, | |
| "learning_rate": 7.696021026559849e-07, | |
| "loss": 0.8505, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0888252148997135, | |
| "grad_norm": 4.085589351210672, | |
| "learning_rate": 7.427392310726087e-07, | |
| "loss": 0.7964, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.146131805157593, | |
| "grad_norm": 3.9343561192796934, | |
| "learning_rate": 7.149259942624286e-07, | |
| "loss": 0.7918, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.146131805157593, | |
| "eval_loss": 0.8878653049468994, | |
| "eval_runtime": 42.3514, | |
| "eval_samples_per_second": 58.534, | |
| "eval_steps_per_second": 0.921, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.2034383954154728, | |
| "grad_norm": 3.882747185049188, | |
| "learning_rate": 6.862712857610811e-07, | |
| "loss": 0.7851, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.2607449856733524, | |
| "grad_norm": 3.9752866239432185, | |
| "learning_rate": 6.568872936084788e-07, | |
| "loss": 0.7941, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.3180515759312321, | |
| "grad_norm": 3.6828284446292505, | |
| "learning_rate": 6.26889061113621e-07, | |
| "loss": 0.7818, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.3753581661891117, | |
| "grad_norm": 4.0779987976559795, | |
| "learning_rate": 5.963940364405425e-07, | |
| "loss": 0.7888, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.4326647564469914, | |
| "grad_norm": 3.658488328256016, | |
| "learning_rate": 5.655216127788472e-07, | |
| "loss": 0.7783, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.4899713467048712, | |
| "grad_norm": 3.9469181853717594, | |
| "learning_rate": 5.343926608991379e-07, | |
| "loss": 0.7718, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5472779369627507, | |
| "grad_norm": 3.9580241692624294, | |
| "learning_rate": 5.031290559234649e-07, | |
| "loss": 0.7789, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.6045845272206303, | |
| "grad_norm": 3.9554511858811114, | |
| "learning_rate": 4.718532001635686e-07, | |
| "loss": 0.7749, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.66189111747851, | |
| "grad_norm": 3.91874392198526, | |
| "learning_rate": 4.406875438950861e-07, | |
| "loss": 0.7639, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.7191977077363898, | |
| "grad_norm": 3.775379766157052, | |
| "learning_rate": 4.097541059439698e-07, | |
| "loss": 0.7632, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.7191977077363898, | |
| "eval_loss": 0.8642120361328125, | |
| "eval_runtime": 42.3256, | |
| "eval_samples_per_second": 58.57, | |
| "eval_steps_per_second": 0.921, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.7765042979942693, | |
| "grad_norm": 3.9001768665312837, | |
| "learning_rate": 3.7917399596210535e-07, | |
| "loss": 0.7686, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.8338108882521489, | |
| "grad_norm": 3.776759348608646, | |
| "learning_rate": 3.490669402625007e-07, | |
| "loss": 0.7652, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.8911174785100286, | |
| "grad_norm": 3.954259341942418, | |
| "learning_rate": 3.195508130704795e-07, | |
| "loss": 0.7608, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.9484240687679084, | |
| "grad_norm": 3.9914749419039275, | |
| "learning_rate": 2.9074117502611296e-07, | |
| "loss": 0.7559, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.005730659025788, | |
| "grad_norm": 4.489725135115642, | |
| "learning_rate": 2.6275082074473075e-07, | |
| "loss": 0.7395, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.0630372492836675, | |
| "grad_norm": 4.317747898535887, | |
| "learning_rate": 2.3568933720688543e-07, | |
| "loss": 0.6774, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.1203438395415475, | |
| "grad_norm": 4.302108159226055, | |
| "learning_rate": 2.096626747067527e-07, | |
| "loss": 0.6677, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.177650429799427, | |
| "grad_norm": 4.307581231308638, | |
| "learning_rate": 1.8477273203877398e-07, | |
| "loss": 0.6876, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.2349570200573066, | |
| "grad_norm": 4.196370522955929, | |
| "learning_rate": 1.6111695754660664e-07, | |
| "loss": 0.6838, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.292263610315186, | |
| "grad_norm": 4.075486625054549, | |
| "learning_rate": 1.3878796759634542e-07, | |
| "loss": 0.6847, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.292263610315186, | |
| "eval_loss": 0.8614758253097534, | |
| "eval_runtime": 42.3657, | |
| "eval_samples_per_second": 58.514, | |
| "eval_steps_per_second": 0.921, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.349570200573066, | |
| "grad_norm": 4.204097889941997, | |
| "learning_rate": 1.1787318396775186e-07, | |
| "loss": 0.6856, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 2.4068767908309456, | |
| "grad_norm": 4.419337399898955, | |
| "learning_rate": 9.845449158317215e-08, | |
| "loss": 0.6915, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.464183381088825, | |
| "grad_norm": 4.134150727696896, | |
| "learning_rate": 8.060791791418886e-08, | |
| "loss": 0.6697, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 2.5214899713467047, | |
| "grad_norm": 4.168834465021675, | |
| "learning_rate": 6.440333532118503e-08, | |
| "loss": 0.6839, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.5787965616045847, | |
| "grad_norm": 4.273918640205208, | |
| "learning_rate": 4.990418749121178e-08, | |
| "loss": 0.6739, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.6361031518624642, | |
| "grad_norm": 4.307832832639697, | |
| "learning_rate": 3.716724104520247e-08, | |
| "loss": 0.6754, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.693409742120344, | |
| "grad_norm": 4.0666737832982385, | |
| "learning_rate": 2.624236328703061e-08, | |
| "loss": 0.6954, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 2.7507163323782233, | |
| "grad_norm": 3.9813588727036855, | |
| "learning_rate": 1.7172326964564775e-08, | |
| "loss": 0.6716, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.8080229226361033, | |
| "grad_norm": 4.208145042622316, | |
| "learning_rate": 9.992642807111484e-09, | |
| "loss": 0.6893, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 2.865329512893983, | |
| "grad_norm": 4.214969116203108, | |
| "learning_rate": 4.7314204948923354e-09, | |
| "loss": 0.685, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.865329512893983, | |
| "eval_loss": 0.8606014847755432, | |
| "eval_runtime": 42.4569, | |
| "eval_samples_per_second": 58.389, | |
| "eval_steps_per_second": 0.919, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.9226361031518624, | |
| "grad_norm": 4.130008937743228, | |
| "learning_rate": 1.4092586048820575e-09, | |
| "loss": 0.6778, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 2.9799426934097424, | |
| "grad_norm": 4.115719334147374, | |
| "learning_rate": 3.91639638886998e-11, | |
| "loss": 0.6874, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.9914040114613183, | |
| "step": 261, | |
| "total_flos": 1538740350615552.0, | |
| "train_loss": 0.8155917972897204, | |
| "train_runtime": 3563.3225, | |
| "train_samples_per_second": 18.784, | |
| "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 | |
| } | |