| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.997342781222321, | |
| "eval_steps": 500, | |
| "global_step": 1692, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.2252470254898071, | |
| "learning_rate": 4.411764705882353e-06, | |
| "loss": 2.746, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 1.2458932399749756, | |
| "learning_rate": 8.823529411764707e-06, | |
| "loss": 2.7375, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 0.9806848764419556, | |
| "learning_rate": 1.323529411764706e-05, | |
| "loss": 2.6088, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.7579270601272583, | |
| "learning_rate": 1.7647058823529414e-05, | |
| "loss": 2.3883, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 0.5973591804504395, | |
| "learning_rate": 2.2058823529411766e-05, | |
| "loss": 2.148, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.4005584120750427, | |
| "learning_rate": 2.647058823529412e-05, | |
| "loss": 1.9697, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 0.30321258306503296, | |
| "learning_rate": 2.990144546649146e-05, | |
| "loss": 1.8807, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 0.33162617683410645, | |
| "learning_rate": 2.9408672798948752e-05, | |
| "loss": 1.788, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.35974565148353577, | |
| "learning_rate": 2.8915900131406044e-05, | |
| "loss": 1.7825, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 0.33461177349090576, | |
| "learning_rate": 2.842312746386334e-05, | |
| "loss": 1.7241, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 0.31440624594688416, | |
| "learning_rate": 2.793035479632063e-05, | |
| "loss": 1.6708, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 0.32463395595550537, | |
| "learning_rate": 2.7437582128777926e-05, | |
| "loss": 1.6098, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 0.35138121247291565, | |
| "learning_rate": 2.6944809461235218e-05, | |
| "loss": 1.5992, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 0.3213985562324524, | |
| "learning_rate": 2.645203679369251e-05, | |
| "loss": 1.5133, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.30699512362480164, | |
| "learning_rate": 2.5959264126149805e-05, | |
| "loss": 1.4807, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.3282407522201538, | |
| "learning_rate": 2.5466491458607097e-05, | |
| "loss": 1.4928, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.36608022451400757, | |
| "learning_rate": 2.4973718791064392e-05, | |
| "loss": 1.4802, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.38351428508758545, | |
| "learning_rate": 2.4480946123521684e-05, | |
| "loss": 1.4886, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.3251235783100128, | |
| "learning_rate": 2.3988173455978975e-05, | |
| "loss": 1.4557, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.3361181914806366, | |
| "learning_rate": 2.349540078843627e-05, | |
| "loss": 1.3857, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 0.36659279465675354, | |
| "learning_rate": 2.3002628120893562e-05, | |
| "loss": 1.4257, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.4663073420524597, | |
| "learning_rate": 2.2509855453350857e-05, | |
| "loss": 1.4516, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 0.3494812846183777, | |
| "learning_rate": 2.201708278580815e-05, | |
| "loss": 1.3714, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 0.3586205542087555, | |
| "learning_rate": 2.152431011826544e-05, | |
| "loss": 1.3569, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 0.38619765639305115, | |
| "learning_rate": 2.1031537450722736e-05, | |
| "loss": 1.3331, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.3586406111717224, | |
| "learning_rate": 2.0538764783180025e-05, | |
| "loss": 1.3446, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.33255332708358765, | |
| "learning_rate": 2.004599211563732e-05, | |
| "loss": 1.3421, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.362377792596817, | |
| "learning_rate": 1.955321944809461e-05, | |
| "loss": 1.3543, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 0.43340814113616943, | |
| "learning_rate": 1.9060446780551903e-05, | |
| "loss": 1.3592, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 0.43384671211242676, | |
| "learning_rate": 1.85676741130092e-05, | |
| "loss": 1.3393, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 0.6346977353096008, | |
| "learning_rate": 1.807490144546649e-05, | |
| "loss": 1.3142, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 0.37973758578300476, | |
| "learning_rate": 1.7582128777923785e-05, | |
| "loss": 1.3279, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 0.3827611207962036, | |
| "learning_rate": 1.7089356110381077e-05, | |
| "loss": 1.3263, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 0.4215668737888336, | |
| "learning_rate": 1.659658344283837e-05, | |
| "loss": 1.3353, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 0.3658856153488159, | |
| "learning_rate": 1.6103810775295664e-05, | |
| "loss": 1.3144, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 0.3511466383934021, | |
| "learning_rate": 1.5611038107752956e-05, | |
| "loss": 1.2621, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 0.3667345941066742, | |
| "learning_rate": 1.511826544021025e-05, | |
| "loss": 1.2563, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 0.40708670020103455, | |
| "learning_rate": 1.4625492772667543e-05, | |
| "loss": 1.2945, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 0.42138373851776123, | |
| "learning_rate": 1.4132720105124836e-05, | |
| "loss": 1.3062, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 0.35037311911582947, | |
| "learning_rate": 1.3639947437582128e-05, | |
| "loss": 1.2745, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 0.48709043860435486, | |
| "learning_rate": 1.3147174770039422e-05, | |
| "loss": 1.3324, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 0.38190776109695435, | |
| "learning_rate": 1.2654402102496715e-05, | |
| "loss": 1.2465, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.4355185329914093, | |
| "learning_rate": 1.2161629434954009e-05, | |
| "loss": 1.2555, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 0.41925591230392456, | |
| "learning_rate": 1.16688567674113e-05, | |
| "loss": 1.2977, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 0.3973592221736908, | |
| "learning_rate": 1.1176084099868594e-05, | |
| "loss": 1.2695, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 0.35202836990356445, | |
| "learning_rate": 1.0683311432325887e-05, | |
| "loss": 1.3174, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.40330058336257935, | |
| "learning_rate": 1.019053876478318e-05, | |
| "loss": 1.2708, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 0.37632250785827637, | |
| "learning_rate": 9.697766097240474e-06, | |
| "loss": 1.2483, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 0.3991636633872986, | |
| "learning_rate": 9.204993429697766e-06, | |
| "loss": 1.2635, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 0.400860071182251, | |
| "learning_rate": 8.71222076215506e-06, | |
| "loss": 1.2464, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 0.40742385387420654, | |
| "learning_rate": 8.219448094612353e-06, | |
| "loss": 1.2532, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 0.4327584207057953, | |
| "learning_rate": 7.726675427069646e-06, | |
| "loss": 1.2771, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 0.3787235915660858, | |
| "learning_rate": 7.233902759526938e-06, | |
| "loss": 1.2508, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 0.423367977142334, | |
| "learning_rate": 6.741130091984232e-06, | |
| "loss": 1.2382, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "grad_norm": 0.3932174742221832, | |
| "learning_rate": 6.248357424441524e-06, | |
| "loss": 1.2759, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.40788665413856506, | |
| "learning_rate": 5.755584756898818e-06, | |
| "loss": 1.2471, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "grad_norm": 0.456950843334198, | |
| "learning_rate": 5.26281208935611e-06, | |
| "loss": 1.2384, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 0.36899831891059875, | |
| "learning_rate": 4.770039421813404e-06, | |
| "loss": 1.1969, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 0.44063258171081543, | |
| "learning_rate": 4.277266754270697e-06, | |
| "loss": 1.2758, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "grad_norm": 0.3583914339542389, | |
| "learning_rate": 3.784494086727989e-06, | |
| "loss": 1.2332, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 0.4087056517601013, | |
| "learning_rate": 3.2917214191852826e-06, | |
| "loss": 1.2646, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "grad_norm": 0.44617870450019836, | |
| "learning_rate": 2.7989487516425756e-06, | |
| "loss": 1.2265, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 0.4349097013473511, | |
| "learning_rate": 2.3061760840998687e-06, | |
| "loss": 1.2976, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 0.4503318965435028, | |
| "learning_rate": 1.8134034165571617e-06, | |
| "loss": 1.2467, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.40000948309898376, | |
| "learning_rate": 1.3206307490144548e-06, | |
| "loss": 1.2146, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "grad_norm": 0.39695021510124207, | |
| "learning_rate": 8.278580814717477e-07, | |
| "loss": 1.263, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 0.4137335419654846, | |
| "learning_rate": 3.3508541392904077e-07, | |
| "loss": 1.2436, | |
| "step": 1675 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 1692, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 7269909176254464.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |