| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 711, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0, | |
| "grad_norm": 6.7530055713569785, | |
| "learning_rate": 2.7777777777777776e-07, | |
| "loss": 1.4781, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 6.606048584660583, | |
| "learning_rate": 1.3888888888888892e-06, | |
| "loss": 1.3861, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 5.138983961367188, | |
| "learning_rate": 2.7777777777777783e-06, | |
| "loss": 1.4422, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 1.2528549295189984, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 1.3677, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 1.6826024664234687, | |
| "learning_rate": 5.555555555555557e-06, | |
| "loss": 1.3277, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 1.049153253140281, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 1.3513, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 1.1637754450319215, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 1.3602, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 0.9614901239317104, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 1.3316, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 0.8800168840313798, | |
| "learning_rate": 1.1111111111111113e-05, | |
| "loss": 1.2865, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 0.8743147668040568, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.3439, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 0.8090973361916171, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 1.3323, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.7791934698691907, | |
| "learning_rate": 1.5277777777777777e-05, | |
| "loss": 1.3316, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.7332468554264387, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 1.3418, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.7367875442634256, | |
| "learning_rate": 1.8055555555555558e-05, | |
| "loss": 1.3488, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 3.454271660640428, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 1.3402, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.41303669264439, | |
| "learning_rate": 1.999891231617599e-05, | |
| "loss": 1.3198, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.7111760423501922, | |
| "learning_rate": 1.9992266216318037e-05, | |
| "loss": 1.291, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.6985734110465796, | |
| "learning_rate": 1.997958229642588e-05, | |
| "loss": 1.3195, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 0.6774596159630265, | |
| "learning_rate": 1.996086822074945e-05, | |
| "loss": 1.3296, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.68714971160374, | |
| "learning_rate": 1.9936135297256183e-05, | |
| "loss": 1.3106, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 0.7055100697390134, | |
| "learning_rate": 1.9905398470798207e-05, | |
| "loss": 1.3251, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.7442856779901117, | |
| "learning_rate": 1.9868676314081907e-05, | |
| "loss": 1.3154, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.6786674709866586, | |
| "learning_rate": 1.9825991016445387e-05, | |
| "loss": 1.305, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.701320710348796, | |
| "learning_rate": 1.9777368370450582e-05, | |
| "loss": 1.2797, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 0.7694290958620503, | |
| "learning_rate": 1.9722837756298112e-05, | |
| "loss": 1.3094, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.7198163251696982, | |
| "learning_rate": 1.9662432124074325e-05, | |
| "loss": 1.3283, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.7451781825097182, | |
| "learning_rate": 1.9596187973841218e-05, | |
| "loss": 1.2966, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.7153678487864736, | |
| "learning_rate": 1.9524145333581315e-05, | |
| "loss": 1.3192, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 2.8370131166342687, | |
| "learning_rate": 1.9446347735010765e-05, | |
| "loss": 1.2988, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.7273784346266812, | |
| "learning_rate": 1.9362842187275354e-05, | |
| "loss": 1.2815, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 0.7610942375358005, | |
| "learning_rate": 1.9273679148545246e-05, | |
| "loss": 1.3121, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 0.719303870712118, | |
| "learning_rate": 1.917891249552568e-05, | |
| "loss": 1.3196, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.6786014692189536, | |
| "learning_rate": 1.9078599490901984e-05, | |
| "loss": 1.3033, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 0.6704522856064519, | |
| "learning_rate": 1.897280074873868e-05, | |
| "loss": 1.2896, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.6769924161484794, | |
| "learning_rate": 1.8861580197853423e-05, | |
| "loss": 1.2808, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 0.7274733092656871, | |
| "learning_rate": 1.8745005043188104e-05, | |
| "loss": 1.3052, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 0.7143035239809652, | |
| "learning_rate": 1.862314572520028e-05, | |
| "loss": 1.3003, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 0.7146392109064191, | |
| "learning_rate": 1.8496075877299585e-05, | |
| "loss": 1.2748, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.7061207940556439, | |
| "learning_rate": 1.8363872281354796e-05, | |
| "loss": 1.305, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.6945437033981094, | |
| "learning_rate": 1.8226614821298444e-05, | |
| "loss": 1.3006, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.6847371880670944, | |
| "learning_rate": 1.808438643485698e-05, | |
| "loss": 1.2705, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.6784008869899278, | |
| "learning_rate": 1.793727306343574e-05, | |
| "loss": 1.2532, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.7037870009631069, | |
| "learning_rate": 1.7785363600188894e-05, | |
| "loss": 1.2638, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.6664930383710185, | |
| "learning_rate": 1.762874983630582e-05, | |
| "loss": 1.2586, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.6773413303705628, | |
| "learning_rate": 1.7467526405546344e-05, | |
| "loss": 1.3138, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 0.6935109696579653, | |
| "learning_rate": 1.7301790727058344e-05, | |
| "loss": 1.2752, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.7212086786238647, | |
| "learning_rate": 1.7131642946512314e-05, | |
| "loss": 1.3091, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 0.7924454572717287, | |
| "learning_rate": 1.6957185875588403e-05, | |
| "loss": 1.3015, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.2710920572280884, | |
| "eval_runtime": 4.2577, | |
| "eval_samples_per_second": 346.195, | |
| "eval_steps_per_second": 5.637, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 0.796377689560899, | |
| "learning_rate": 1.6778524929852513e-05, | |
| "loss": 1.2628, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 0.7364886186224426, | |
| "learning_rate": 1.659576806505905e-05, | |
| "loss": 1.2284, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.7049213011696083, | |
| "learning_rate": 1.640902571191869e-05, | |
| "loss": 1.2187, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 1.9377903386730986, | |
| "learning_rate": 1.6218410709370735e-05, | |
| "loss": 1.2438, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 0.7011828464498495, | |
| "learning_rate": 1.6024038236400246e-05, | |
| "loss": 1.1998, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 2.5996069923788436, | |
| "learning_rate": 1.582602574244121e-05, | |
| "loss": 1.1919, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 0.679979871549876, | |
| "learning_rate": 1.562449287640781e-05, | |
| "loss": 1.2456, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.71927352282914, | |
| "learning_rate": 1.5419561414396657e-05, | |
| "loss": 1.2493, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 0.7190913662714845, | |
| "learning_rate": 1.5211355186103655e-05, | |
| "loss": 1.2058, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.7182017182110072, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 1.2371, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 0.705180903431495, | |
| "learning_rate": 1.4785623567312492e-05, | |
| "loss": 1.1976, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.6883033004979775, | |
| "learning_rate": 1.4568355424854113e-05, | |
| "loss": 1.2768, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 0.6815574246451964, | |
| "learning_rate": 1.4348326856751496e-05, | |
| "loss": 1.2334, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 0.6804999351991194, | |
| "learning_rate": 1.412567081511659e-05, | |
| "loss": 1.2183, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 0.7193921850993512, | |
| "learning_rate": 1.3900521839710428e-05, | |
| "loss": 1.2032, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 0.6954927887572464, | |
| "learning_rate": 1.367301597664757e-05, | |
| "loss": 1.2289, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 0.708546670689236, | |
| "learning_rate": 1.3443290696190335e-05, | |
| "loss": 1.229, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 0.7124305795556933, | |
| "learning_rate": 1.3211484809682482e-05, | |
| "loss": 1.2434, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 0.6866449286630997, | |
| "learning_rate": 1.2977738385672558e-05, | |
| "loss": 1.2587, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 0.7158743341618767, | |
| "learning_rate": 1.2742192665277569e-05, | |
| "loss": 1.2522, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 0.6741832926767023, | |
| "learning_rate": 1.250498997683813e-05, | |
| "loss": 1.2318, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 0.7018162451875303, | |
| "learning_rate": 1.2266273649916669e-05, | |
| "loss": 1.2017, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.6814770297513967, | |
| "learning_rate": 1.202618792869063e-05, | |
| "loss": 1.2043, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.6877213251148805, | |
| "learning_rate": 1.178487788479303e-05, | |
| "loss": 1.216, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 0.6722566603288663, | |
| "learning_rate": 1.1542489329653024e-05, | |
| "loss": 1.2367, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 0.6835971864599367, | |
| "learning_rate": 1.129916872638945e-05, | |
| "loss": 1.2611, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.7040578619734443, | |
| "learning_rate": 1.1055063101310581e-05, | |
| "loss": 1.2252, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 0.679881563899439, | |
| "learning_rate": 1.08103199550736e-05, | |
| "loss": 1.2321, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.6794312105046864, | |
| "learning_rate": 1.0565087173557396e-05, | |
| "loss": 1.2252, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 0.6889618373928458, | |
| "learning_rate": 1.0319512938502654e-05, | |
| "loss": 1.2256, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 0.6921906751981153, | |
| "learning_rate": 1.0073745637973125e-05, | |
| "loss": 1.2344, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 0.6993543043577237, | |
| "learning_rate": 9.827933776692236e-06, | |
| "loss": 1.2391, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 0.6959462039565184, | |
| "learning_rate": 9.582225886309217e-06, | |
| "loss": 1.2155, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 0.6863551427574576, | |
| "learning_rate": 9.336770435648963e-06, | |
| "loss": 1.252, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 0.7184772923029549, | |
| "learning_rate": 9.091715740999829e-06, | |
| "loss": 1.2295, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 0.6949962185725445, | |
| "learning_rate": 8.84720987649363e-06, | |
| "loss": 1.2044, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 0.6710588422516326, | |
| "learning_rate": 8.60340058463194e-06, | |
| "loss": 1.2407, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 0.6482064194341762, | |
| "learning_rate": 8.360435187012789e-06, | |
| "loss": 1.2637, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 0.6715553002740511, | |
| "learning_rate": 8.118460495311687e-06, | |
| "loss": 1.2795, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 1.1764818802442565, | |
| "learning_rate": 7.877622722570772e-06, | |
| "loss": 1.2228, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 0.6932612325260208, | |
| "learning_rate": 7.638067394849672e-06, | |
| "loss": 1.2272, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.7017921905179951, | |
| "learning_rate": 7.3999392632914936e-06, | |
| "loss": 1.2476, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.6556538474805375, | |
| "learning_rate": 7.163382216657033e-06, | |
| "loss": 1.2249, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 0.6649783838358192, | |
| "learning_rate": 6.928539194380101e-06, | |
| "loss": 1.2166, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 0.6582704106972542, | |
| "learning_rate": 6.6955521001964526e-06, | |
| "loss": 1.2149, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.6842373684969378, | |
| "learning_rate": 6.464561716398565e-06, | |
| "loss": 1.2089, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 0.6763842453630291, | |
| "learning_rate": 6.2357076187680325e-06, | |
| "loss": 1.2352, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.2562438249588013, | |
| "eval_runtime": 4.0229, | |
| "eval_samples_per_second": 366.403, | |
| "eval_steps_per_second": 5.966, | |
| "step": 474 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.703770585858162, | |
| "learning_rate": 6.009128092236983e-06, | |
| "loss": 1.2048, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 0.6786892775253128, | |
| "learning_rate": 5.78496004732952e-06, | |
| "loss": 1.1518, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 0.70326378053169, | |
| "learning_rate": 5.563338937433622e-06, | |
| "loss": 1.197, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 0.6916875060651468, | |
| "learning_rate": 5.344398676953526e-06, | |
| "loss": 1.2111, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 0.6475152204206446, | |
| "learning_rate": 5.128271560392037e-06, | |
| "loss": 1.1739, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 0.6813942164053138, | |
| "learning_rate": 4.915088182411674e-06, | |
| "loss": 1.1905, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 0.6486602612694303, | |
| "learning_rate": 4.7049773589229306e-06, | |
| "loss": 1.2063, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 0.6731868552171755, | |
| "learning_rate": 4.498066049247344e-06, | |
| "loss": 1.2198, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 0.685788252482629, | |
| "learning_rate": 4.29447927940242e-06, | |
| "loss": 1.1706, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 0.6713780690925405, | |
| "learning_rate": 4.094340066554742e-06, | |
| "loss": 1.1887, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 0.6669181028932704, | |
| "learning_rate": 3.897769344686929e-06, | |
| "loss": 1.1933, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.6681179890022899, | |
| "learning_rate": 3.7048858915233665e-06, | |
| "loss": 1.2061, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 0.6688074058150556, | |
| "learning_rate": 3.5158062567588468e-06, | |
| "loss": 1.1585, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 0.6500738740113599, | |
| "learning_rate": 3.330644691633492e-06, | |
| "loss": 1.1793, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.6546313173839559, | |
| "learning_rate": 3.149513079896521e-06, | |
| "loss": 1.2088, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.6544093029161898, | |
| "learning_rate": 2.9725208702005736e-06, | |
| "loss": 1.1975, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "grad_norm": 0.6458002129205276, | |
| "learning_rate": 2.7997750099674282e-06, | |
| "loss": 1.1913, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "grad_norm": 0.6571017394804074, | |
| "learning_rate": 2.631379880765107e-06, | |
| "loss": 1.1876, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "grad_norm": 0.6563720392014831, | |
| "learning_rate": 2.467437235235378e-06, | |
| "loss": 1.1677, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "grad_norm": 0.6640825824608834, | |
| "learning_rate": 2.3080461356097938e-06, | |
| "loss": 1.1876, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 0.6332904461492966, | |
| "learning_rate": 2.153302893851401e-06, | |
| "loss": 1.1553, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "grad_norm": 0.654023164181366, | |
| "learning_rate": 2.0033010134583085e-06, | |
| "loss": 1.1896, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 0.6359350507881623, | |
| "learning_rate": 1.8581311329642592e-06, | |
| "loss": 1.1526, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 0.6238559177863219, | |
| "learning_rate": 1.7178809711703525e-06, | |
| "loss": 1.1847, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "grad_norm": 0.6479836099852851, | |
| "learning_rate": 1.5826352741410333e-06, | |
| "loss": 1.1883, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 0.6615350392909103, | |
| "learning_rate": 1.452475763996326e-06, | |
| "loss": 1.1562, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 0.659005508964398, | |
| "learning_rate": 1.3274810895313083e-06, | |
| "loss": 1.1666, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 0.6530124794164938, | |
| "learning_rate": 1.207726778692625e-06, | |
| "loss": 1.1694, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "grad_norm": 0.637671650897288, | |
| "learning_rate": 1.0932851929407828e-06, | |
| "loss": 1.21, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "grad_norm": 0.6426650375048046, | |
| "learning_rate": 9.842254835257792e-07, | |
| "loss": 1.1983, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.6549427233533902, | |
| "learning_rate": 8.806135497025181e-07, | |
| "loss": 1.1796, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 0.637483284013688, | |
| "learning_rate": 7.825119989112173e-07, | |
| "loss": 1.2032, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 0.6458643762516288, | |
| "learning_rate": 6.899801089469205e-07, | |
| "loss": 1.1898, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 0.6560437400254089, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 1.1815, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 0.6346340189110924, | |
| "learning_rate": 5.218455615757601e-07, | |
| "loss": 1.1829, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 0.6408145596532006, | |
| "learning_rate": 4.4634449935427203e-07, | |
| "loss": 1.1653, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "grad_norm": 0.6553954626425612, | |
| "learning_rate": 3.7661622694171393e-07, | |
| "loss": 1.2129, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 0.6949305164155417, | |
| "learning_rate": 3.127028775990515e-07, | |
| "loss": 1.1822, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 0.6657650684680331, | |
| "learning_rate": 2.546430709239578e-07, | |
| "loss": 1.1818, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 0.6438354912565324, | |
| "learning_rate": 2.02471889514948e-07, | |
| "loss": 1.1696, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.6330124825001247, | |
| "learning_rate": 1.562208577727442e-07, | |
| "loss": 1.1443, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 0.6389378157102353, | |
| "learning_rate": 1.1591792285167603e-07, | |
| "loss": 1.1657, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 0.6353907126618263, | |
| "learning_rate": 8.158743777263334e-08, | |
| "loss": 1.183, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 0.644204938538006, | |
| "learning_rate": 5.325014670776951e-08, | |
| "loss": 1.1641, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "grad_norm": 0.6585816956828645, | |
| "learning_rate": 3.092317244584919e-08, | |
| "loss": 1.2134, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 0.6523630752795228, | |
| "learning_rate": 1.4620006045816814e-08, | |
| "loss": 1.2174, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 0.659207596228058, | |
| "learning_rate": 4.3504986848297295e-09, | |
| "loss": 1.1732, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.6428302138785816, | |
| "learning_rate": 1.2085570569642102e-10, | |
| "loss": 1.1719, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.2565690279006958, | |
| "eval_runtime": 4.0115, | |
| "eval_samples_per_second": 367.448, | |
| "eval_steps_per_second": 5.983, | |
| "step": 711 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 711, | |
| "total_flos": 138562624290816.0, | |
| "train_loss": 1.242754081298195, | |
| "train_runtime": 610.9223, | |
| "train_samples_per_second": 148.462, | |
| "train_steps_per_second": 1.164 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 711, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 138562624290816.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |