| { | |
| "best_metric": 0.31647276878356934, | |
| "best_model_checkpoint": "deberta-v3-xsmall-zyda-2-transformed-quality-new/checkpoint-37947", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 37947, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03952881650723377, | |
| "grad_norm": 2.9192252159118652, | |
| "learning_rate": 4.9341186391546106e-05, | |
| "loss": 0.6141, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07905763301446754, | |
| "grad_norm": 2.9683828353881836, | |
| "learning_rate": 4.868237278309221e-05, | |
| "loss": 0.5164, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11858644952170132, | |
| "grad_norm": 2.587739944458008, | |
| "learning_rate": 4.802355917463832e-05, | |
| "loss": 0.4719, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.15811526602893508, | |
| "grad_norm": 2.8187127113342285, | |
| "learning_rate": 4.7364745566184415e-05, | |
| "loss": 0.457, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.19764408253616886, | |
| "grad_norm": 3.325552463531494, | |
| "learning_rate": 4.670593195773052e-05, | |
| "loss": 0.4503, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.23717289904340264, | |
| "grad_norm": 2.489327907562256, | |
| "learning_rate": 4.604711834927662e-05, | |
| "loss": 0.4345, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.27670171555063644, | |
| "grad_norm": 3.6023690700531006, | |
| "learning_rate": 4.538830474082273e-05, | |
| "loss": 0.4185, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.31623053205787016, | |
| "grad_norm": 2.4426589012145996, | |
| "learning_rate": 4.4729491132368835e-05, | |
| "loss": 0.4095, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.35575934856510394, | |
| "grad_norm": 3.424283981323242, | |
| "learning_rate": 4.407067752391494e-05, | |
| "loss": 0.399, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.3952881650723377, | |
| "grad_norm": 2.923548936843872, | |
| "learning_rate": 4.3411863915461035e-05, | |
| "loss": 0.3984, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.4348169815795715, | |
| "grad_norm": 2.6173856258392334, | |
| "learning_rate": 4.2753050307007145e-05, | |
| "loss": 0.3896, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.47434579808680527, | |
| "grad_norm": 2.0746352672576904, | |
| "learning_rate": 4.209423669855325e-05, | |
| "loss": 0.3809, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.513874614594039, | |
| "grad_norm": 1.683587908744812, | |
| "learning_rate": 4.143542309009935e-05, | |
| "loss": 0.3805, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.5534034311012729, | |
| "grad_norm": 3.254350185394287, | |
| "learning_rate": 4.0776609481645455e-05, | |
| "loss": 0.3761, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.5929322476085066, | |
| "grad_norm": 2.623117685317993, | |
| "learning_rate": 4.011779587319156e-05, | |
| "loss": 0.3806, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.6324610641157403, | |
| "grad_norm": 2.246891975402832, | |
| "learning_rate": 3.945898226473767e-05, | |
| "loss": 0.3609, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.6719898806229742, | |
| "grad_norm": 3.263880729675293, | |
| "learning_rate": 3.8800168656283764e-05, | |
| "loss": 0.3717, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.7115186971302079, | |
| "grad_norm": 2.449371337890625, | |
| "learning_rate": 3.814135504782987e-05, | |
| "loss": 0.3693, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.7510475136374417, | |
| "grad_norm": 2.3527493476867676, | |
| "learning_rate": 3.748254143937597e-05, | |
| "loss": 0.3623, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.7905763301446754, | |
| "grad_norm": 2.8773348331451416, | |
| "learning_rate": 3.682372783092208e-05, | |
| "loss": 0.3615, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.8301051466519093, | |
| "grad_norm": 2.7280490398406982, | |
| "learning_rate": 3.6164914222468184e-05, | |
| "loss": 0.3598, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.869633963159143, | |
| "grad_norm": 2.2904133796691895, | |
| "learning_rate": 3.550610061401429e-05, | |
| "loss": 0.3539, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.9091627796663768, | |
| "grad_norm": 2.197935104370117, | |
| "learning_rate": 3.4847287005560384e-05, | |
| "loss": 0.3563, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.9486915961736105, | |
| "grad_norm": 2.6248714923858643, | |
| "learning_rate": 3.4188473397106494e-05, | |
| "loss": 0.353, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.9882204126808444, | |
| "grad_norm": 2.276596784591675, | |
| "learning_rate": 3.35296597886526e-05, | |
| "loss": 0.3506, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.35121020674705505, | |
| "eval_mse": 0.35121019676996473, | |
| "eval_runtime": 148.1236, | |
| "eval_samples_per_second": 675.112, | |
| "eval_steps_per_second": 84.389, | |
| "step": 12649 | |
| }, | |
| { | |
| "epoch": 1.027749229188078, | |
| "grad_norm": 2.749833822250366, | |
| "learning_rate": 3.28708461801987e-05, | |
| "loss": 0.3211, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.067278045695312, | |
| "grad_norm": 2.4262359142303467, | |
| "learning_rate": 3.2212032571744803e-05, | |
| "loss": 0.303, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.1068068622025455, | |
| "grad_norm": 2.7832248210906982, | |
| "learning_rate": 3.155321896329091e-05, | |
| "loss": 0.3047, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.1463356787097794, | |
| "grad_norm": 2.4965128898620605, | |
| "learning_rate": 3.089440535483701e-05, | |
| "loss": 0.2994, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.1858644952170132, | |
| "grad_norm": 2.200387716293335, | |
| "learning_rate": 3.0235591746383113e-05, | |
| "loss": 0.3, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.225393311724247, | |
| "grad_norm": 2.688816547393799, | |
| "learning_rate": 2.9576778137929216e-05, | |
| "loss": 0.2993, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.2649221282314809, | |
| "grad_norm": 3.349059820175171, | |
| "learning_rate": 2.8917964529475323e-05, | |
| "loss": 0.297, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.3044509447387145, | |
| "grad_norm": 2.224276542663574, | |
| "learning_rate": 2.8259150921021426e-05, | |
| "loss": 0.2986, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.3439797612459483, | |
| "grad_norm": 2.157297134399414, | |
| "learning_rate": 2.7600337312567533e-05, | |
| "loss": 0.2978, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.3835085777531821, | |
| "grad_norm": 2.9185595512390137, | |
| "learning_rate": 2.6941523704113636e-05, | |
| "loss": 0.2964, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.4230373942604158, | |
| "grad_norm": 2.66011381149292, | |
| "learning_rate": 2.6282710095659736e-05, | |
| "loss": 0.2931, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.4625662107676496, | |
| "grad_norm": 3.047621011734009, | |
| "learning_rate": 2.562389648720584e-05, | |
| "loss": 0.2941, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.5020950272748834, | |
| "grad_norm": 2.8634958267211914, | |
| "learning_rate": 2.4965082878751946e-05, | |
| "loss": 0.2933, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.541623843782117, | |
| "grad_norm": 2.827955722808838, | |
| "learning_rate": 2.430626927029805e-05, | |
| "loss": 0.2909, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.581152660289351, | |
| "grad_norm": 2.969160556793213, | |
| "learning_rate": 2.3647455661844152e-05, | |
| "loss": 0.2892, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.6206814767965847, | |
| "grad_norm": 2.9614064693450928, | |
| "learning_rate": 2.2988642053390256e-05, | |
| "loss": 0.2897, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.6602102933038185, | |
| "grad_norm": 2.6613972187042236, | |
| "learning_rate": 2.232982844493636e-05, | |
| "loss": 0.2886, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.6997391098110524, | |
| "grad_norm": 2.762899398803711, | |
| "learning_rate": 2.1671014836482465e-05, | |
| "loss": 0.2872, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.739267926318286, | |
| "grad_norm": 3.2839698791503906, | |
| "learning_rate": 2.1012201228028565e-05, | |
| "loss": 0.2895, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.7787967428255198, | |
| "grad_norm": 2.6103265285491943, | |
| "learning_rate": 2.0353387619574672e-05, | |
| "loss": 0.2872, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.8183255593327536, | |
| "grad_norm": 3.014410972595215, | |
| "learning_rate": 1.9694574011120775e-05, | |
| "loss": 0.2902, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.8578543758399872, | |
| "grad_norm": 3.351289987564087, | |
| "learning_rate": 1.903576040266688e-05, | |
| "loss": 0.2876, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.897383192347221, | |
| "grad_norm": 3.1844663619995117, | |
| "learning_rate": 1.837694679421298e-05, | |
| "loss": 0.2855, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.936912008854455, | |
| "grad_norm": 2.6804494857788086, | |
| "learning_rate": 1.7718133185759085e-05, | |
| "loss": 0.2859, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 1.9764408253616885, | |
| "grad_norm": 2.4572935104370117, | |
| "learning_rate": 1.7059319577305188e-05, | |
| "loss": 0.28, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.3186613619327545, | |
| "eval_mse": 0.31866135950110924, | |
| "eval_runtime": 145.5886, | |
| "eval_samples_per_second": 686.867, | |
| "eval_steps_per_second": 85.858, | |
| "step": 25298 | |
| }, | |
| { | |
| "epoch": 2.0159696418689226, | |
| "grad_norm": 2.0255794525146484, | |
| "learning_rate": 1.640050596885129e-05, | |
| "loss": 0.2658, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 2.055498458376156, | |
| "grad_norm": 2.699540853500366, | |
| "learning_rate": 1.5741692360397398e-05, | |
| "loss": 0.2477, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.09502727488339, | |
| "grad_norm": 3.3947625160217285, | |
| "learning_rate": 1.5082878751943503e-05, | |
| "loss": 0.2466, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 2.134556091390624, | |
| "grad_norm": 2.3115530014038086, | |
| "learning_rate": 1.4424065143489604e-05, | |
| "loss": 0.2465, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.1740849078978575, | |
| "grad_norm": 2.9776363372802734, | |
| "learning_rate": 1.376525153503571e-05, | |
| "loss": 0.2446, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 2.213613724405091, | |
| "grad_norm": 1.9914076328277588, | |
| "learning_rate": 1.3106437926581813e-05, | |
| "loss": 0.2417, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.253142540912325, | |
| "grad_norm": 2.153264284133911, | |
| "learning_rate": 1.2447624318127916e-05, | |
| "loss": 0.2436, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 2.2926713574195587, | |
| "grad_norm": 2.458815574645996, | |
| "learning_rate": 1.1788810709674019e-05, | |
| "loss": 0.2446, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.332200173926793, | |
| "grad_norm": 2.1611809730529785, | |
| "learning_rate": 1.1129997101220122e-05, | |
| "loss": 0.2408, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.3717289904340264, | |
| "grad_norm": 3.270725727081299, | |
| "learning_rate": 1.0471183492766227e-05, | |
| "loss": 0.2446, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.41125780694126, | |
| "grad_norm": 2.7944698333740234, | |
| "learning_rate": 9.81236988431233e-06, | |
| "loss": 0.242, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.450786623448494, | |
| "grad_norm": 2.138493299484253, | |
| "learning_rate": 9.153556275858434e-06, | |
| "loss": 0.2428, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.4903154399557277, | |
| "grad_norm": 2.3900632858276367, | |
| "learning_rate": 8.494742667404539e-06, | |
| "loss": 0.2435, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.5298442564629617, | |
| "grad_norm": 3.044337749481201, | |
| "learning_rate": 7.835929058950642e-06, | |
| "loss": 0.2411, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.5693730729701953, | |
| "grad_norm": 2.801241159439087, | |
| "learning_rate": 7.177115450496746e-06, | |
| "loss": 0.2437, | |
| "step": 32500 | |
| }, | |
| { | |
| "epoch": 2.608901889477429, | |
| "grad_norm": 2.554879903793335, | |
| "learning_rate": 6.518301842042849e-06, | |
| "loss": 0.2436, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.648430705984663, | |
| "grad_norm": 2.7090861797332764, | |
| "learning_rate": 5.859488233588953e-06, | |
| "loss": 0.2403, | |
| "step": 33500 | |
| }, | |
| { | |
| "epoch": 2.6879595224918966, | |
| "grad_norm": 2.0840466022491455, | |
| "learning_rate": 5.200674625135057e-06, | |
| "loss": 0.2396, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.7274883389991302, | |
| "grad_norm": 2.8196446895599365, | |
| "learning_rate": 4.541861016681161e-06, | |
| "loss": 0.2372, | |
| "step": 34500 | |
| }, | |
| { | |
| "epoch": 2.7670171555063643, | |
| "grad_norm": 2.4318623542785645, | |
| "learning_rate": 3.883047408227265e-06, | |
| "loss": 0.2399, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.806545972013598, | |
| "grad_norm": 3.474367380142212, | |
| "learning_rate": 3.2242337997733685e-06, | |
| "loss": 0.2368, | |
| "step": 35500 | |
| }, | |
| { | |
| "epoch": 2.8460747885208315, | |
| "grad_norm": 2.736551523208618, | |
| "learning_rate": 2.565420191319472e-06, | |
| "loss": 0.2367, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.8856036050280656, | |
| "grad_norm": 1.705733060836792, | |
| "learning_rate": 1.9066065828655756e-06, | |
| "loss": 0.2397, | |
| "step": 36500 | |
| }, | |
| { | |
| "epoch": 2.925132421535299, | |
| "grad_norm": 2.315380811691284, | |
| "learning_rate": 1.2477929744116795e-06, | |
| "loss": 0.2388, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 2.964661238042533, | |
| "grad_norm": 2.4691686630249023, | |
| "learning_rate": 5.889793659577832e-07, | |
| "loss": 0.2398, | |
| "step": 37500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.31647276878356934, | |
| "eval_mse": 0.3164727601450849, | |
| "eval_runtime": 132.665, | |
| "eval_samples_per_second": 753.778, | |
| "eval_steps_per_second": 94.222, | |
| "step": 37947 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 37947, | |
| "total_flos": 3.999532597714253e+16, | |
| "train_loss": 0.31243033600999526, | |
| "train_runtime": 5395.7257, | |
| "train_samples_per_second": 450.097, | |
| "train_steps_per_second": 7.033 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 37947, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.999532597714253e+16, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |