| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 59.7710843373494, | |
| "eval_steps": 10, | |
| "global_step": 60, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7710843373493976, | |
| "grad_norm": 3.914616675051536, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.6485056281089783, | |
| "memory(GiB)": 76.1, | |
| "step": 1, | |
| "token_acc": 0.8288129939443704, | |
| "train_speed(iter/s)": 0.00248 | |
| }, | |
| { | |
| "epoch": 1.7710843373493976, | |
| "grad_norm": 7.135952401058273, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.306803822517395, | |
| "memory(GiB)": 77.59, | |
| "step": 2, | |
| "token_acc": 0.8355539032171605, | |
| "train_speed(iter/s)": 0.002291 | |
| }, | |
| { | |
| "epoch": 2.7710843373493974, | |
| "grad_norm": 7.25317611851469, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.2851154804229736, | |
| "memory(GiB)": 77.59, | |
| "step": 3, | |
| "token_acc": 0.834915035973237, | |
| "train_speed(iter/s)": 0.002247 | |
| }, | |
| { | |
| "epoch": 3.7710843373493974, | |
| "grad_norm": 6.865015999508222, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.290574073791504, | |
| "memory(GiB)": 77.59, | |
| "step": 4, | |
| "token_acc": 0.8338767344065504, | |
| "train_speed(iter/s)": 0.002214 | |
| }, | |
| { | |
| "epoch": 4.771084337349397, | |
| "grad_norm": 6.57767604145361, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.261496901512146, | |
| "memory(GiB)": 77.59, | |
| "step": 5, | |
| "token_acc": 0.8348027461456405, | |
| "train_speed(iter/s)": 0.0022 | |
| }, | |
| { | |
| "epoch": 5.771084337349397, | |
| "grad_norm": 5.131561904948073, | |
| "learning_rate": 3e-06, | |
| "loss": 1.20093834400177, | |
| "memory(GiB)": 77.59, | |
| "step": 6, | |
| "token_acc": 0.8447540089861448, | |
| "train_speed(iter/s)": 0.002191 | |
| }, | |
| { | |
| "epoch": 6.771084337349397, | |
| "grad_norm": 4.169361922366313, | |
| "learning_rate": 3.5e-06, | |
| "loss": 1.118795394897461, | |
| "memory(GiB)": 77.59, | |
| "step": 7, | |
| "token_acc": 0.845426595733518, | |
| "train_speed(iter/s)": 0.002187 | |
| }, | |
| { | |
| "epoch": 7.771084337349397, | |
| "grad_norm": 4.513043229540615, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.0338046550750732, | |
| "memory(GiB)": 77.59, | |
| "step": 8, | |
| "token_acc": 0.8524168712340995, | |
| "train_speed(iter/s)": 0.00218 | |
| }, | |
| { | |
| "epoch": 8.771084337349398, | |
| "grad_norm": 5.185455486257246, | |
| "learning_rate": 4.5e-06, | |
| "loss": 0.9289287328720093, | |
| "memory(GiB)": 77.59, | |
| "step": 9, | |
| "token_acc": 0.8699944903581267, | |
| "train_speed(iter/s)": 0.002177 | |
| }, | |
| { | |
| "epoch": 9.771084337349398, | |
| "grad_norm": 3.6878315103563324, | |
| "learning_rate": 5e-06, | |
| "loss": 0.9007519483566284, | |
| "memory(GiB)": 77.59, | |
| "step": 10, | |
| "token_acc": 0.8636064441638881, | |
| "train_speed(iter/s)": 0.002176 | |
| }, | |
| { | |
| "epoch": 10.771084337349398, | |
| "grad_norm": 3.153907065444448, | |
| "learning_rate": 4.999658262481173e-06, | |
| "loss": 0.8321974277496338, | |
| "memory(GiB)": 77.59, | |
| "step": 11, | |
| "token_acc": 0.8710321094284695, | |
| "train_speed(iter/s)": 0.002152 | |
| }, | |
| { | |
| "epoch": 11.771084337349398, | |
| "grad_norm": 2.203676742176422, | |
| "learning_rate": 4.998633143352315e-06, | |
| "loss": 0.7872496247291565, | |
| "memory(GiB)": 77.59, | |
| "step": 12, | |
| "token_acc": 0.8761502671032225, | |
| "train_speed(iter/s)": 0.002149 | |
| }, | |
| { | |
| "epoch": 12.771084337349398, | |
| "grad_norm": 2.4327146254784084, | |
| "learning_rate": 4.9969249228707625e-06, | |
| "loss": 0.7419657707214355, | |
| "memory(GiB)": 77.59, | |
| "step": 13, | |
| "token_acc": 0.8786807478287534, | |
| "train_speed(iter/s)": 0.002148 | |
| }, | |
| { | |
| "epoch": 13.771084337349398, | |
| "grad_norm": 1.8937402046126173, | |
| "learning_rate": 4.994534068046936e-06, | |
| "loss": 0.709729790687561, | |
| "memory(GiB)": 77.59, | |
| "step": 14, | |
| "token_acc": 0.8879921788137868, | |
| "train_speed(iter/s)": 0.002151 | |
| }, | |
| { | |
| "epoch": 14.771084337349398, | |
| "grad_norm": 1.4897225411877277, | |
| "learning_rate": 4.991461232516675e-06, | |
| "loss": 0.676007866859436, | |
| "memory(GiB)": 77.59, | |
| "step": 15, | |
| "token_acc": 0.8962045012503473, | |
| "train_speed(iter/s)": 0.00215 | |
| }, | |
| { | |
| "epoch": 15.771084337349398, | |
| "grad_norm": 1.297539359495755, | |
| "learning_rate": 4.987707256362529e-06, | |
| "loss": 0.6474949717521667, | |
| "memory(GiB)": 77.59, | |
| "step": 16, | |
| "token_acc": 0.8947659474239368, | |
| "train_speed(iter/s)": 0.002148 | |
| }, | |
| { | |
| "epoch": 16.771084337349397, | |
| "grad_norm": 0.7199351026372981, | |
| "learning_rate": 4.983273165884096e-06, | |
| "loss": 0.6222354173660278, | |
| "memory(GiB)": 77.59, | |
| "step": 17, | |
| "token_acc": 0.9017643862580863, | |
| "train_speed(iter/s)": 0.00215 | |
| }, | |
| { | |
| "epoch": 17.771084337349397, | |
| "grad_norm": 1.4007282443863567, | |
| "learning_rate": 4.978160173317439e-06, | |
| "loss": 0.6050044298171997, | |
| "memory(GiB)": 77.59, | |
| "step": 18, | |
| "token_acc": 0.899228721854176, | |
| "train_speed(iter/s)": 0.00215 | |
| }, | |
| { | |
| "epoch": 18.771084337349397, | |
| "grad_norm": 1.4351431315774181, | |
| "learning_rate": 4.972369676503672e-06, | |
| "loss": 0.5903453230857849, | |
| "memory(GiB)": 77.59, | |
| "step": 19, | |
| "token_acc": 0.9062169786627631, | |
| "train_speed(iter/s)": 0.002149 | |
| }, | |
| { | |
| "epoch": 19.771084337349397, | |
| "grad_norm": 1.1929137455373535, | |
| "learning_rate": 4.965903258506806e-06, | |
| "loss": 0.5688210725784302, | |
| "memory(GiB)": 77.59, | |
| "step": 20, | |
| "token_acc": 0.9079767112253118, | |
| "train_speed(iter/s)": 0.002148 | |
| }, | |
| { | |
| "epoch": 20.771084337349397, | |
| "grad_norm": 1.1732494606890316, | |
| "learning_rate": 4.9587626871809564e-06, | |
| "loss": 0.543393611907959, | |
| "memory(GiB)": 77.59, | |
| "step": 21, | |
| "token_acc": 0.9126149389091825, | |
| "train_speed(iter/s)": 0.002136 | |
| }, | |
| { | |
| "epoch": 21.771084337349397, | |
| "grad_norm": 1.0573555070054292, | |
| "learning_rate": 4.950949914687024e-06, | |
| "loss": 0.5107942223548889, | |
| "memory(GiB)": 77.59, | |
| "step": 22, | |
| "token_acc": 0.913213028660819, | |
| "train_speed(iter/s)": 0.002136 | |
| }, | |
| { | |
| "epoch": 22.771084337349397, | |
| "grad_norm": 0.9119223186948517, | |
| "learning_rate": 4.942467076958999e-06, | |
| "loss": 0.48675835132598877, | |
| "memory(GiB)": 77.59, | |
| "step": 23, | |
| "token_acc": 0.916867432182973, | |
| "train_speed(iter/s)": 0.002137 | |
| }, | |
| { | |
| "epoch": 23.771084337349397, | |
| "grad_norm": 1.0782070696527255, | |
| "learning_rate": 4.933316493120015e-06, | |
| "loss": 0.5053229331970215, | |
| "memory(GiB)": 77.59, | |
| "step": 24, | |
| "token_acc": 0.9218187118672303, | |
| "train_speed(iter/s)": 0.002139 | |
| }, | |
| { | |
| "epoch": 24.771084337349397, | |
| "grad_norm": 0.9655818937470999, | |
| "learning_rate": 4.923500664848327e-06, | |
| "loss": 0.5017877817153931, | |
| "memory(GiB)": 77.59, | |
| "step": 25, | |
| "token_acc": 0.9221839217084671, | |
| "train_speed(iter/s)": 0.002138 | |
| }, | |
| { | |
| "epoch": 25.771084337349397, | |
| "grad_norm": 0.9404583758277602, | |
| "learning_rate": 4.913022275693372e-06, | |
| "loss": 0.4680900275707245, | |
| "memory(GiB)": 77.59, | |
| "step": 26, | |
| "token_acc": 0.9232181127591402, | |
| "train_speed(iter/s)": 0.002139 | |
| }, | |
| { | |
| "epoch": 26.771084337349397, | |
| "grad_norm": 0.9566563796831183, | |
| "learning_rate": 4.901884190342121e-06, | |
| "loss": 0.4565889239311218, | |
| "memory(GiB)": 77.59, | |
| "step": 27, | |
| "token_acc": 0.9250697693507346, | |
| "train_speed(iter/s)": 0.002138 | |
| }, | |
| { | |
| "epoch": 27.771084337349397, | |
| "grad_norm": 0.7703532186007063, | |
| "learning_rate": 4.890089453835894e-06, | |
| "loss": 0.43708479404449463, | |
| "memory(GiB)": 77.59, | |
| "step": 28, | |
| "token_acc": 0.9259928949775574, | |
| "train_speed(iter/s)": 0.002139 | |
| }, | |
| { | |
| "epoch": 28.771084337349397, | |
| "grad_norm": 0.9233409831228016, | |
| "learning_rate": 4.8776412907378845e-06, | |
| "loss": 0.4270290732383728, | |
| "memory(GiB)": 77.59, | |
| "step": 29, | |
| "token_acc": 0.9246415671992716, | |
| "train_speed(iter/s)": 0.00214 | |
| }, | |
| { | |
| "epoch": 29.771084337349397, | |
| "grad_norm": 0.9965352268428146, | |
| "learning_rate": 4.864543104251587e-06, | |
| "loss": 0.41402751207351685, | |
| "memory(GiB)": 77.59, | |
| "step": 30, | |
| "token_acc": 0.9240786009778112, | |
| "train_speed(iter/s)": 0.002139 | |
| }, | |
| { | |
| "epoch": 30.771084337349397, | |
| "grad_norm": 0.9358295902337755, | |
| "learning_rate": 4.850798475290403e-06, | |
| "loss": 0.39457955956459045, | |
| "memory(GiB)": 77.59, | |
| "step": 31, | |
| "token_acc": 0.9368182401505754, | |
| "train_speed(iter/s)": 0.002131 | |
| }, | |
| { | |
| "epoch": 31.771084337349397, | |
| "grad_norm": 0.9813539670443086, | |
| "learning_rate": 4.836411161498653e-06, | |
| "loss": 0.38136690855026245, | |
| "memory(GiB)": 77.59, | |
| "step": 32, | |
| "token_acc": 0.9375963637091769, | |
| "train_speed(iter/s)": 0.002133 | |
| }, | |
| { | |
| "epoch": 32.7710843373494, | |
| "grad_norm": 0.4679577449591144, | |
| "learning_rate": 4.821385096224268e-06, | |
| "loss": 0.36845850944519043, | |
| "memory(GiB)": 77.59, | |
| "step": 33, | |
| "token_acc": 0.9348571791559913, | |
| "train_speed(iter/s)": 0.002134 | |
| }, | |
| { | |
| "epoch": 33.7710843373494, | |
| "grad_norm": 0.9212511288160155, | |
| "learning_rate": 4.8057243874434625e-06, | |
| "loss": 0.35925933718681335, | |
| "memory(GiB)": 77.59, | |
| "step": 34, | |
| "token_acc": 0.9376081366560561, | |
| "train_speed(iter/s)": 0.002133 | |
| }, | |
| { | |
| "epoch": 34.7710843373494, | |
| "grad_norm": 1.0561026403132139, | |
| "learning_rate": 4.789433316637644e-06, | |
| "loss": 0.3401709794998169, | |
| "memory(GiB)": 77.59, | |
| "step": 35, | |
| "token_acc": 0.9403078788403664, | |
| "train_speed(iter/s)": 0.002134 | |
| }, | |
| { | |
| "epoch": 35.7710843373494, | |
| "grad_norm": 0.936417548403036, | |
| "learning_rate": 4.772516337622907e-06, | |
| "loss": 0.3241081237792969, | |
| "memory(GiB)": 77.59, | |
| "step": 36, | |
| "token_acc": 0.9484973388290848, | |
| "train_speed(iter/s)": 0.002135 | |
| }, | |
| { | |
| "epoch": 36.7710843373494, | |
| "grad_norm": 1.116983283196546, | |
| "learning_rate": 4.754978075332398e-06, | |
| "loss": 0.30902814865112305, | |
| "memory(GiB)": 77.59, | |
| "step": 37, | |
| "token_acc": 0.9476456995060315, | |
| "train_speed(iter/s)": 0.002135 | |
| }, | |
| { | |
| "epoch": 37.7710843373494, | |
| "grad_norm": 0.9993160029154087, | |
| "learning_rate": 4.736823324551909e-06, | |
| "loss": 0.308858722448349, | |
| "memory(GiB)": 77.59, | |
| "step": 38, | |
| "token_acc": 0.9504927069901695, | |
| "train_speed(iter/s)": 0.002135 | |
| }, | |
| { | |
| "epoch": 38.7710843373494, | |
| "grad_norm": 0.9122824683879396, | |
| "learning_rate": 4.71805704860903e-06, | |
| "loss": 0.27843528985977173, | |
| "memory(GiB)": 77.59, | |
| "step": 39, | |
| "token_acc": 0.9496681563352376, | |
| "train_speed(iter/s)": 0.002136 | |
| }, | |
| { | |
| "epoch": 39.7710843373494, | |
| "grad_norm": 1.009388916794573, | |
| "learning_rate": 4.698684378016223e-06, | |
| "loss": 0.25660544633865356, | |
| "memory(GiB)": 77.59, | |
| "step": 40, | |
| "token_acc": 0.9552901130798869, | |
| "train_speed(iter/s)": 0.002137 | |
| }, | |
| { | |
| "epoch": 40.7710843373494, | |
| "grad_norm": 1.1895642620511482, | |
| "learning_rate": 4.678710609068193e-06, | |
| "loss": 0.2424250692129135, | |
| "memory(GiB)": 77.59, | |
| "step": 41, | |
| "token_acc": 0.9549429605785101, | |
| "train_speed(iter/s)": 0.002131 | |
| }, | |
| { | |
| "epoch": 41.7710843373494, | |
| "grad_norm": 1.1193041843299223, | |
| "learning_rate": 4.658141202393935e-06, | |
| "loss": 0.23843874037265778, | |
| "memory(GiB)": 77.59, | |
| "step": 42, | |
| "token_acc": 0.9608553608553608, | |
| "train_speed(iter/s)": 0.002131 | |
| }, | |
| { | |
| "epoch": 42.7710843373494, | |
| "grad_norm": 1.110610217839956, | |
| "learning_rate": 4.636981781463848e-06, | |
| "loss": 0.210187166929245, | |
| "memory(GiB)": 77.59, | |
| "step": 43, | |
| "token_acc": 0.9653150186596419, | |
| "train_speed(iter/s)": 0.002132 | |
| }, | |
| { | |
| "epoch": 43.7710843373494, | |
| "grad_norm": 0.8486274696165886, | |
| "learning_rate": 4.615238131052339e-06, | |
| "loss": 0.21901345252990723, | |
| "memory(GiB)": 77.59, | |
| "step": 44, | |
| "token_acc": 0.9653416261042005, | |
| "train_speed(iter/s)": 0.002132 | |
| }, | |
| { | |
| "epoch": 44.7710843373494, | |
| "grad_norm": 1.0423804412798912, | |
| "learning_rate": 4.592916195656322e-06, | |
| "loss": 0.17482870817184448, | |
| "memory(GiB)": 77.59, | |
| "step": 45, | |
| "token_acc": 0.9674520144027672, | |
| "train_speed(iter/s)": 0.002133 | |
| }, | |
| { | |
| "epoch": 45.7710843373494, | |
| "grad_norm": 1.2841486053816782, | |
| "learning_rate": 4.570022077870051e-06, | |
| "loss": 0.18200109899044037, | |
| "memory(GiB)": 77.59, | |
| "step": 46, | |
| "token_acc": 0.9708482572798199, | |
| "train_speed(iter/s)": 0.002133 | |
| }, | |
| { | |
| "epoch": 46.7710843373494, | |
| "grad_norm": 1.172978940255451, | |
| "learning_rate": 4.546562036716732e-06, | |
| "loss": 0.15860611200332642, | |
| "memory(GiB)": 77.59, | |
| "step": 47, | |
| "token_acc": 0.9780445141443995, | |
| "train_speed(iter/s)": 0.002133 | |
| }, | |
| { | |
| "epoch": 47.7710843373494, | |
| "grad_norm": 1.2360197347924713, | |
| "learning_rate": 4.522542485937369e-06, | |
| "loss": 0.14376184344291687, | |
| "memory(GiB)": 77.59, | |
| "step": 48, | |
| "token_acc": 0.9766241102700259, | |
| "train_speed(iter/s)": 0.002135 | |
| }, | |
| { | |
| "epoch": 48.7710843373494, | |
| "grad_norm": 0.6385686989268172, | |
| "learning_rate": 4.497969992237312e-06, | |
| "loss": 0.13206440210342407, | |
| "memory(GiB)": 77.59, | |
| "step": 49, | |
| "token_acc": 0.9818039023693685, | |
| "train_speed(iter/s)": 0.002135 | |
| }, | |
| { | |
| "epoch": 49.7710843373494, | |
| "grad_norm": 1.2901120485113573, | |
| "learning_rate": 4.472851273490985e-06, | |
| "loss": 0.11426319181919098, | |
| "memory(GiB)": 77.59, | |
| "step": 50, | |
| "token_acc": 0.9836425932423786, | |
| "train_speed(iter/s)": 0.002135 | |
| }, | |
| { | |
| "epoch": 50.7710843373494, | |
| "grad_norm": 1.2068228907293697, | |
| "learning_rate": 4.4471931969052816e-06, | |
| "loss": 0.09582371264696121, | |
| "memory(GiB)": 77.59, | |
| "step": 51, | |
| "token_acc": 0.9849937986991469, | |
| "train_speed(iter/s)": 0.00213 | |
| }, | |
| { | |
| "epoch": 51.7710843373494, | |
| "grad_norm": 1.1813520357475888, | |
| "learning_rate": 4.421002777142148e-06, | |
| "loss": 0.09163334965705872, | |
| "memory(GiB)": 77.59, | |
| "step": 52, | |
| "token_acc": 0.985701349843115, | |
| "train_speed(iter/s)": 0.00213 | |
| }, | |
| { | |
| "epoch": 52.7710843373494, | |
| "grad_norm": 1.1122597460598151, | |
| "learning_rate": 4.394287174400838e-06, | |
| "loss": 0.08439977467060089, | |
| "memory(GiB)": 77.59, | |
| "step": 53, | |
| "token_acc": 0.9878925052136355, | |
| "train_speed(iter/s)": 0.00213 | |
| }, | |
| { | |
| "epoch": 53.7710843373494, | |
| "grad_norm": 1.1253515530343927, | |
| "learning_rate": 4.3670536924603855e-06, | |
| "loss": 0.07314425706863403, | |
| "memory(GiB)": 77.59, | |
| "step": 54, | |
| "token_acc": 0.9900777579782926, | |
| "train_speed(iter/s)": 0.00213 | |
| }, | |
| { | |
| "epoch": 54.7710843373494, | |
| "grad_norm": 1.1033670330706447, | |
| "learning_rate": 4.33930977668283e-06, | |
| "loss": 0.06532438099384308, | |
| "memory(GiB)": 77.59, | |
| "step": 55, | |
| "token_acc": 0.9898691656771185, | |
| "train_speed(iter/s)": 0.00213 | |
| }, | |
| { | |
| "epoch": 55.7710843373494, | |
| "grad_norm": 1.1439476751450148, | |
| "learning_rate": 4.311063011977723e-06, | |
| "loss": 0.06318923085927963, | |
| "memory(GiB)": 77.59, | |
| "step": 56, | |
| "token_acc": 0.9903093019267052, | |
| "train_speed(iter/s)": 0.002131 | |
| }, | |
| { | |
| "epoch": 56.7710843373494, | |
| "grad_norm": 1.1421545710607997, | |
| "learning_rate": 4.282321120728493e-06, | |
| "loss": 0.05167176201939583, | |
| "memory(GiB)": 77.59, | |
| "step": 57, | |
| "token_acc": 0.991860976682798, | |
| "train_speed(iter/s)": 0.002131 | |
| }, | |
| { | |
| "epoch": 57.7710843373494, | |
| "grad_norm": 1.1132394753881205, | |
| "learning_rate": 4.253091960681222e-06, | |
| "loss": 0.05162680149078369, | |
| "memory(GiB)": 77.59, | |
| "step": 58, | |
| "token_acc": 0.992666204277479, | |
| "train_speed(iter/s)": 0.002132 | |
| }, | |
| { | |
| "epoch": 58.7710843373494, | |
| "grad_norm": 1.1383077361483038, | |
| "learning_rate": 4.2233835227964145e-06, | |
| "loss": 0.044882968068122864, | |
| "memory(GiB)": 77.59, | |
| "step": 59, | |
| "token_acc": 0.9941307654265692, | |
| "train_speed(iter/s)": 0.002132 | |
| }, | |
| { | |
| "epoch": 59.7710843373494, | |
| "grad_norm": 0.9459729284657843, | |
| "learning_rate": 4.1932039290643534e-06, | |
| "loss": 0.042324937880039215, | |
| "memory(GiB)": 77.59, | |
| "step": 60, | |
| "token_acc": 0.9941203679909544, | |
| "train_speed(iter/s)": 0.002132 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 200, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 268227602284544.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |