| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.995418771290967, | |
| "eval_steps": 500, | |
| "global_step": 25500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 5.873140172278778e-05, | |
| "loss": 0.0966, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 0.03481524437665939, | |
| "eval_runtime": 123.2077, | |
| "eval_samples_per_second": 53.097, | |
| "eval_steps_per_second": 6.639, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00011746280344557555, | |
| "loss": 0.0446, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_loss": 0.038774896413087845, | |
| "eval_runtime": 123.3038, | |
| "eval_samples_per_second": 53.056, | |
| "eval_steps_per_second": 6.634, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00017619420516836332, | |
| "loss": 0.063, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_loss": 0.06299161165952682, | |
| "eval_runtime": 123.374, | |
| "eval_samples_per_second": 53.026, | |
| "eval_steps_per_second": 6.63, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0002349256068911511, | |
| "loss": 0.0938, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "eval_loss": 0.07528574019670486, | |
| "eval_runtime": 123.3537, | |
| "eval_samples_per_second": 53.034, | |
| "eval_steps_per_second": 6.631, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002936570086139389, | |
| "loss": 0.1281, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 0.10252855718135834, | |
| "eval_runtime": 123.3622, | |
| "eval_samples_per_second": 53.031, | |
| "eval_steps_per_second": 6.631, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.00029417881226887095, | |
| "loss": 0.1452, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.12126505374908447, | |
| "eval_runtime": 123.2879, | |
| "eval_samples_per_second": 53.063, | |
| "eval_steps_per_second": 6.635, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002876528170546008, | |
| "loss": 0.1393, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_loss": 0.0877891555428505, | |
| "eval_runtime": 123.3134, | |
| "eval_samples_per_second": 53.052, | |
| "eval_steps_per_second": 6.634, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 0.00028112682184033063, | |
| "loss": 0.1278, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 0.08643390238285065, | |
| "eval_runtime": 123.5525, | |
| "eval_samples_per_second": 52.949, | |
| "eval_steps_per_second": 6.621, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00027460082662606047, | |
| "loss": 0.121, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_loss": 0.07961534708738327, | |
| "eval_runtime": 123.4172, | |
| "eval_samples_per_second": 53.007, | |
| "eval_steps_per_second": 6.628, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00026807483141179025, | |
| "loss": 0.115, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_loss": 0.07281593233346939, | |
| "eval_runtime": 123.3851, | |
| "eval_samples_per_second": 53.021, | |
| "eval_steps_per_second": 6.63, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0002615488361975201, | |
| "loss": 0.1105, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 0.06985253840684891, | |
| "eval_runtime": 123.3894, | |
| "eval_samples_per_second": 53.019, | |
| "eval_steps_per_second": 6.629, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.00025502284098324993, | |
| "loss": 0.109, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_loss": 0.0665203332901001, | |
| "eval_runtime": 123.2402, | |
| "eval_samples_per_second": 53.083, | |
| "eval_steps_per_second": 6.637, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.00024849684576897977, | |
| "loss": 0.0989, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 0.06210066005587578, | |
| "eval_runtime": 123.1909, | |
| "eval_samples_per_second": 53.105, | |
| "eval_steps_per_second": 6.64, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 0.00024197085055470958, | |
| "loss": 0.0981, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_loss": 0.06805345416069031, | |
| "eval_runtime": 123.2672, | |
| "eval_samples_per_second": 53.072, | |
| "eval_steps_per_second": 6.636, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.0002354448553404394, | |
| "loss": 0.0913, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 0.05965917557477951, | |
| "eval_runtime": 123.3122, | |
| "eval_samples_per_second": 53.052, | |
| "eval_steps_per_second": 6.634, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00022891886012616923, | |
| "loss": 0.0878, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 0.05600811913609505, | |
| "eval_runtime": 123.3189, | |
| "eval_samples_per_second": 53.049, | |
| "eval_steps_per_second": 6.633, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00022239286491189904, | |
| "loss": 0.0852, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.05495968833565712, | |
| "eval_runtime": 123.3222, | |
| "eval_samples_per_second": 53.048, | |
| "eval_steps_per_second": 6.633, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00021586686969762888, | |
| "loss": 0.0708, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_loss": 0.05065647512674332, | |
| "eval_runtime": 123.1018, | |
| "eval_samples_per_second": 53.143, | |
| "eval_steps_per_second": 6.645, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.0002093408744833587, | |
| "loss": 0.0669, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 0.05243635177612305, | |
| "eval_runtime": 123.1404, | |
| "eval_samples_per_second": 53.126, | |
| "eval_steps_per_second": 6.643, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00020281487926908853, | |
| "loss": 0.0657, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "eval_loss": 0.04499243199825287, | |
| "eval_runtime": 123.3227, | |
| "eval_samples_per_second": 53.048, | |
| "eval_steps_per_second": 6.633, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00019628888405481834, | |
| "loss": 0.0636, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_loss": 0.04722798988223076, | |
| "eval_runtime": 123.4099, | |
| "eval_samples_per_second": 53.01, | |
| "eval_steps_per_second": 6.628, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00018976288884054818, | |
| "loss": 0.0623, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 0.042462971061468124, | |
| "eval_runtime": 123.343, | |
| "eval_samples_per_second": 53.039, | |
| "eval_steps_per_second": 6.632, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00018323689362627802, | |
| "loss": 0.0596, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "eval_loss": 0.04393870010972023, | |
| "eval_runtime": 123.1794, | |
| "eval_samples_per_second": 53.11, | |
| "eval_steps_per_second": 6.641, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.00017671089841200783, | |
| "loss": 0.0612, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_loss": 0.03987164422869682, | |
| "eval_runtime": 123.4328, | |
| "eval_samples_per_second": 53.001, | |
| "eval_steps_per_second": 6.627, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 0.00017018490319773767, | |
| "loss": 0.0553, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_loss": 0.04043687880039215, | |
| "eval_runtime": 123.5215, | |
| "eval_samples_per_second": 52.962, | |
| "eval_steps_per_second": 6.622, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00016365890798346748, | |
| "loss": 0.0565, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 0.04023285582661629, | |
| "eval_runtime": 123.3607, | |
| "eval_samples_per_second": 53.031, | |
| "eval_steps_per_second": 6.631, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.00015713291276919726, | |
| "loss": 0.0541, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "eval_loss": 0.03617456555366516, | |
| "eval_runtime": 123.7139, | |
| "eval_samples_per_second": 52.88, | |
| "eval_steps_per_second": 6.612, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.0001506069175549271, | |
| "loss": 0.0527, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "eval_loss": 0.036614831537008286, | |
| "eval_runtime": 123.4987, | |
| "eval_samples_per_second": 52.972, | |
| "eval_steps_per_second": 6.624, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.00014408092234065694, | |
| "loss": 0.0485, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 0.03705143555998802, | |
| "eval_runtime": 123.4413, | |
| "eval_samples_per_second": 52.997, | |
| "eval_steps_per_second": 6.627, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00013755492712638678, | |
| "loss": 0.0502, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": 0.03249647840857506, | |
| "eval_runtime": 123.5764, | |
| "eval_samples_per_second": 52.939, | |
| "eval_steps_per_second": 6.619, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.0001310289319121166, | |
| "loss": 0.0485, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "eval_loss": 0.03329641371965408, | |
| "eval_runtime": 123.7232, | |
| "eval_samples_per_second": 52.876, | |
| "eval_steps_per_second": 6.612, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00012450293669784643, | |
| "loss": 0.0459, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_loss": 0.03429277986288071, | |
| "eval_runtime": 123.8157, | |
| "eval_samples_per_second": 52.837, | |
| "eval_steps_per_second": 6.607, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00011797694148357624, | |
| "loss": 0.0461, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_loss": 0.02981030009686947, | |
| "eval_runtime": 123.4645, | |
| "eval_samples_per_second": 52.987, | |
| "eval_steps_per_second": 6.625, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00011145094626930605, | |
| "loss": 0.0423, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.029031969606876373, | |
| "eval_runtime": 123.8223, | |
| "eval_samples_per_second": 52.834, | |
| "eval_steps_per_second": 6.606, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 0.00010492495105503587, | |
| "loss": 0.0304, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "eval_loss": 0.029610687866806984, | |
| "eval_runtime": 123.3939, | |
| "eval_samples_per_second": 53.017, | |
| "eval_steps_per_second": 6.629, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 9.83989558407657e-05, | |
| "loss": 0.0286, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 0.029453950002789497, | |
| "eval_runtime": 123.7963, | |
| "eval_samples_per_second": 52.845, | |
| "eval_steps_per_second": 6.608, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "learning_rate": 9.187296062649552e-05, | |
| "loss": 0.0294, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_loss": 0.028666863217949867, | |
| "eval_runtime": 123.5343, | |
| "eval_samples_per_second": 52.957, | |
| "eval_steps_per_second": 6.622, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 8.534696541222536e-05, | |
| "loss": 0.0271, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_loss": 0.029929010197520256, | |
| "eval_runtime": 123.4149, | |
| "eval_samples_per_second": 53.008, | |
| "eval_steps_per_second": 6.628, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 7.882097019795519e-05, | |
| "loss": 0.0256, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "eval_loss": 0.027137087658047676, | |
| "eval_runtime": 123.7129, | |
| "eval_samples_per_second": 52.88, | |
| "eval_steps_per_second": 6.612, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 7.229497498368501e-05, | |
| "loss": 0.0256, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "eval_loss": 0.027056274935603142, | |
| "eval_runtime": 123.4616, | |
| "eval_samples_per_second": 52.988, | |
| "eval_steps_per_second": 6.626, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 6.576897976941483e-05, | |
| "loss": 0.0231, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "eval_loss": 0.02597939409315586, | |
| "eval_runtime": 123.5327, | |
| "eval_samples_per_second": 52.958, | |
| "eval_steps_per_second": 6.622, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 5.924298455514466e-05, | |
| "loss": 0.0246, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "eval_loss": 0.025122959166765213, | |
| "eval_runtime": 123.806, | |
| "eval_samples_per_second": 52.841, | |
| "eval_steps_per_second": 6.607, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 5.2716989340874485e-05, | |
| "loss": 0.0235, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "eval_loss": 0.024449240416288376, | |
| "eval_runtime": 123.6218, | |
| "eval_samples_per_second": 52.919, | |
| "eval_steps_per_second": 6.617, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 4.61909941266043e-05, | |
| "loss": 0.0234, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_loss": 0.024168582633137703, | |
| "eval_runtime": 123.8732, | |
| "eval_samples_per_second": 52.812, | |
| "eval_steps_per_second": 6.604, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "learning_rate": 3.966499891233413e-05, | |
| "loss": 0.0224, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "eval_loss": 0.023941034451127052, | |
| "eval_runtime": 123.519, | |
| "eval_samples_per_second": 52.963, | |
| "eval_steps_per_second": 6.622, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 3.313900369806395e-05, | |
| "loss": 0.0214, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_loss": 0.02283557504415512, | |
| "eval_runtime": 123.5762, | |
| "eval_samples_per_second": 52.939, | |
| "eval_steps_per_second": 6.619, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 2.6613008483793777e-05, | |
| "loss": 0.0213, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "eval_loss": 0.0223745945841074, | |
| "eval_runtime": 123.8829, | |
| "eval_samples_per_second": 52.808, | |
| "eval_steps_per_second": 6.603, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 2.0087013269523602e-05, | |
| "loss": 0.0196, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_loss": 0.02261008322238922, | |
| "eval_runtime": 123.6333, | |
| "eval_samples_per_second": 52.915, | |
| "eval_steps_per_second": 6.616, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 1.3561018055253423e-05, | |
| "loss": 0.0194, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_loss": 0.021464873105287552, | |
| "eval_runtime": 124.0332, | |
| "eval_samples_per_second": 52.744, | |
| "eval_steps_per_second": 6.595, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "learning_rate": 7.03502284098325e-06, | |
| "loss": 0.0191, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.94, | |
| "eval_loss": 0.021281694993376732, | |
| "eval_runtime": 123.6857, | |
| "eval_samples_per_second": 52.892, | |
| "eval_steps_per_second": 6.614, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 5.090276267130737e-07, | |
| "loss": 0.0197, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.0211211945861578, | |
| "eval_runtime": 123.7008, | |
| "eval_samples_per_second": 52.886, | |
| "eval_steps_per_second": 6.613, | |
| "step": 25500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 25539, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.428545951977472e+16, | |
| "train_batch_size": 18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |