| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 1548, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 3.802351951599121, | |
| "learning_rate": 3.2258064516129034e-05, | |
| "loss": 3.7331, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 3.442800760269165, | |
| "learning_rate": 6.451612903225807e-05, | |
| "loss": 2.9581, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 3.616957902908325, | |
| "learning_rate": 9.677419354838711e-05, | |
| "loss": 2.4309, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 2.3956120014190674, | |
| "learning_rate": 0.00012903225806451613, | |
| "loss": 2.2667, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 2.698212146759033, | |
| "learning_rate": 0.00016129032258064516, | |
| "loss": 2.0782, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 2.7342419624328613, | |
| "learning_rate": 0.00019354838709677422, | |
| "loss": 2.0691, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 2.039422035217285, | |
| "learning_rate": 0.00019712849964106247, | |
| "loss": 2.0865, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 2.0298664569854736, | |
| "learning_rate": 0.00019353912419239053, | |
| "loss": 1.9552, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 1.8978090286254883, | |
| "learning_rate": 0.0001899497487437186, | |
| "loss": 2.0111, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.9588110446929932, | |
| "learning_rate": 0.00018636037329504667, | |
| "loss": 1.961, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 2.160839796066284, | |
| "learning_rate": 0.00018277099784637474, | |
| "loss": 2.0133, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 2.0899839401245117, | |
| "learning_rate": 0.00017918162239770278, | |
| "loss": 1.958, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 2.029879570007324, | |
| "learning_rate": 0.00017559224694903088, | |
| "loss": 1.9116, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.7021583318710327, | |
| "learning_rate": 0.00017200287150035895, | |
| "loss": 1.8842, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.5238200426101685, | |
| "learning_rate": 0.00016841349605168702, | |
| "loss": 1.8229, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.9168670177459717, | |
| "learning_rate": 0.0001648241206030151, | |
| "loss": 1.9235, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 1.8892511129379272, | |
| "learning_rate": 0.00016123474515434316, | |
| "loss": 1.8859, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 1.7955598831176758, | |
| "learning_rate": 0.00015764536970567123, | |
| "loss": 1.8619, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 1.7658586502075195, | |
| "learning_rate": 0.00015405599425699927, | |
| "loss": 1.8235, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 1.598111629486084, | |
| "learning_rate": 0.00015046661880832737, | |
| "loss": 1.8554, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 1.9352959394454956, | |
| "learning_rate": 0.00014687724335965544, | |
| "loss": 1.801, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 1.9746991395950317, | |
| "learning_rate": 0.0001432878679109835, | |
| "loss": 1.895, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 1.476921796798706, | |
| "learning_rate": 0.00013969849246231157, | |
| "loss": 1.7978, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.667672872543335, | |
| "learning_rate": 0.00013610911701363964, | |
| "loss": 1.76, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 1.5855716466903687, | |
| "learning_rate": 0.00013251974156496769, | |
| "loss": 1.8171, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 1.9429579973220825, | |
| "learning_rate": 0.00012893036611629576, | |
| "loss": 1.811, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 1.684810996055603, | |
| "learning_rate": 0.00012534099066762382, | |
| "loss": 1.8148, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.6180702447891235, | |
| "learning_rate": 0.00012175161521895191, | |
| "loss": 1.7356, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 1.7795788049697876, | |
| "learning_rate": 0.00011816223977027998, | |
| "loss": 1.7969, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 1.7645013332366943, | |
| "learning_rate": 0.00011457286432160806, | |
| "loss": 1.8198, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.713512897491455, | |
| "learning_rate": 0.00011098348887293613, | |
| "loss": 1.7514, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 1.3435418605804443, | |
| "learning_rate": 0.00010739411342426417, | |
| "loss": 1.6679, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 1.5819846391677856, | |
| "learning_rate": 0.00010380473797559225, | |
| "loss": 1.7182, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 1.3324297666549683, | |
| "learning_rate": 0.00010021536252692032, | |
| "loss": 1.7301, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 1.4800920486450195, | |
| "learning_rate": 9.662598707824839e-05, | |
| "loss": 1.7123, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 1.5812880992889404, | |
| "learning_rate": 9.303661162957645e-05, | |
| "loss": 1.7816, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.5376070737838745, | |
| "learning_rate": 8.944723618090453e-05, | |
| "loss": 1.7446, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 1.5884149074554443, | |
| "learning_rate": 8.58578607322326e-05, | |
| "loss": 1.6483, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 2.03753662109375, | |
| "learning_rate": 8.226848528356066e-05, | |
| "loss": 1.6695, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 1.7924336194992065, | |
| "learning_rate": 7.867910983488873e-05, | |
| "loss": 1.7109, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 1.80973219871521, | |
| "learning_rate": 7.508973438621681e-05, | |
| "loss": 1.7017, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 1.7763901948928833, | |
| "learning_rate": 7.150035893754488e-05, | |
| "loss": 1.6669, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 3.0561070442199707, | |
| "learning_rate": 6.791098348887293e-05, | |
| "loss": 1.6681, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 1.7292817831039429, | |
| "learning_rate": 6.4321608040201e-05, | |
| "loss": 1.6787, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 1.6157281398773193, | |
| "learning_rate": 6.073223259152908e-05, | |
| "loss": 1.6716, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 1.5873336791992188, | |
| "learning_rate": 5.714285714285714e-05, | |
| "loss": 1.6724, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 1.7574703693389893, | |
| "learning_rate": 5.355348169418522e-05, | |
| "loss": 1.6608, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 1.7600945234298706, | |
| "learning_rate": 4.996410624551328e-05, | |
| "loss": 1.7204, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 1.4515677690505981, | |
| "learning_rate": 4.6374730796841356e-05, | |
| "loss": 1.6755, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 1.9057530164718628, | |
| "learning_rate": 4.278535534816942e-05, | |
| "loss": 1.6953, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 1.7482775449752808, | |
| "learning_rate": 3.919597989949749e-05, | |
| "loss": 1.686, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 1.8991056680679321, | |
| "learning_rate": 3.560660445082556e-05, | |
| "loss": 1.6611, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 1.7882860898971558, | |
| "learning_rate": 3.201722900215363e-05, | |
| "loss": 1.7103, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 1.6628391742706299, | |
| "learning_rate": 2.84278535534817e-05, | |
| "loss": 1.6925, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 1.5516228675842285, | |
| "learning_rate": 2.4838478104809766e-05, | |
| "loss": 1.652, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 1.5719884634017944, | |
| "learning_rate": 2.1249102656137835e-05, | |
| "loss": 1.6383, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 1.8628472089767456, | |
| "learning_rate": 1.76597272074659e-05, | |
| "loss": 1.6922, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 1.7590694427490234, | |
| "learning_rate": 1.407035175879397e-05, | |
| "loss": 1.687, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 1.6741101741790771, | |
| "learning_rate": 1.048097631012204e-05, | |
| "loss": 1.6599, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 1.7378610372543335, | |
| "learning_rate": 6.891600861450108e-06, | |
| "loss": 1.6765, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "grad_norm": 1.5435367822647095, | |
| "learning_rate": 3.3022254127781766e-06, | |
| "loss": 1.607, | |
| "step": 1525 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1548, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 1.0151913205845197e+17, | |
| "train_batch_size": 6, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |