| { |
| "best_metric": 1.3676984310150146, |
| "best_model_checkpoint": "/mnt/d/f0_kmeans_T5_resume/checkpoint-8000", |
| "epoch": 0.8033287936888481, |
| "eval_steps": 2000, |
| "global_step": 8000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.010041609921110603, |
| "grad_norm": 0.6225447654724121, |
| "learning_rate": 2e-05, |
| "loss": 1.8574, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.020083219842221205, |
| "grad_norm": 0.58293217420578, |
| "learning_rate": 4e-05, |
| "loss": 1.8565, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.030124829763331806, |
| "grad_norm": 0.7323440313339233, |
| "learning_rate": 6e-05, |
| "loss": 1.8463, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.04016643968444241, |
| "grad_norm": 0.501658022403717, |
| "learning_rate": 8e-05, |
| "loss": 1.8284, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.05020804960555301, |
| "grad_norm": 0.971801221370697, |
| "learning_rate": 0.0001, |
| "loss": 1.8107, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.06024965952666361, |
| "grad_norm": 0.6389101147651672, |
| "learning_rate": 9.894269401564813e-05, |
| "loss": 1.7857, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.07029126944777421, |
| "grad_norm": 0.6149382591247559, |
| "learning_rate": 9.788538803129626e-05, |
| "loss": 1.7686, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.08033287936888482, |
| "grad_norm": 0.9680651426315308, |
| "learning_rate": 9.68280820469444e-05, |
| "loss": 1.7481, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.09037448928999542, |
| "grad_norm": 0.5966058969497681, |
| "learning_rate": 9.577077606259251e-05, |
| "loss": 1.7298, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.10041609921110602, |
| "grad_norm": 0.806626558303833, |
| "learning_rate": 9.471347007824066e-05, |
| "loss": 1.7173, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.11045770913221663, |
| "grad_norm": 0.6925487518310547, |
| "learning_rate": 9.365616409388877e-05, |
| "loss": 1.6905, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.12049931905332723, |
| "grad_norm": 0.7050909399986267, |
| "learning_rate": 9.25988581095369e-05, |
| "loss": 1.679, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.13054092897443784, |
| "grad_norm": 0.560664713382721, |
| "learning_rate": 9.154155212518503e-05, |
| "loss": 1.6691, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.14058253889554842, |
| "grad_norm": 0.7096071243286133, |
| "learning_rate": 9.048424614083316e-05, |
| "loss": 1.6542, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.15062414881665903, |
| "grad_norm": 0.8381897211074829, |
| "learning_rate": 8.942694015648129e-05, |
| "loss": 1.6545, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.16066575873776964, |
| "grad_norm": 0.8832806944847107, |
| "learning_rate": 8.836963417212942e-05, |
| "loss": 1.6378, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.17070736865888023, |
| "grad_norm": 0.5661273002624512, |
| "learning_rate": 8.731232818777755e-05, |
| "loss": 1.625, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.18074897857999084, |
| "grad_norm": 0.4389723241329193, |
| "learning_rate": 8.625502220342567e-05, |
| "loss": 1.6253, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.19079058850110145, |
| "grad_norm": 0.45324140787124634, |
| "learning_rate": 8.519771621907381e-05, |
| "loss": 1.6048, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.20083219842221203, |
| "grad_norm": 0.6926701068878174, |
| "learning_rate": 8.414041023472193e-05, |
| "loss": 1.6103, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.20083219842221203, |
| "eval_loss": 1.4620566368103027, |
| "eval_runtime": 407.2516, |
| "eval_samples_per_second": 173.89, |
| "eval_steps_per_second": 21.738, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.21087380834332264, |
| "grad_norm": 0.5207750797271729, |
| "learning_rate": 8.308310425037006e-05, |
| "loss": 1.6101, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.22091541826443326, |
| "grad_norm": 0.564079225063324, |
| "learning_rate": 8.202579826601819e-05, |
| "loss": 1.5943, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.23095702818554384, |
| "grad_norm": 0.6724056601524353, |
| "learning_rate": 8.096849228166632e-05, |
| "loss": 1.5963, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.24099863810665445, |
| "grad_norm": 0.42919212579727173, |
| "learning_rate": 7.991118629731445e-05, |
| "loss": 1.5863, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.25104024802776503, |
| "grad_norm": 0.49685391783714294, |
| "learning_rate": 7.885388031296258e-05, |
| "loss": 1.5833, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.2610818579488757, |
| "grad_norm": 0.784831702709198, |
| "learning_rate": 7.779657432861071e-05, |
| "loss": 1.5759, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.27112346786998626, |
| "grad_norm": 0.3932148814201355, |
| "learning_rate": 7.673926834425883e-05, |
| "loss": 1.5764, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.28116507779109684, |
| "grad_norm": 0.7318882346153259, |
| "learning_rate": 7.568196235990697e-05, |
| "loss": 1.5531, |
| "step": 2800 |
| }, |
| { |
| "epoch": 0.2912066877122075, |
| "grad_norm": 0.46883219480514526, |
| "learning_rate": 7.462465637555509e-05, |
| "loss": 1.5597, |
| "step": 2900 |
| }, |
| { |
| "epoch": 0.30124829763331806, |
| "grad_norm": 0.5420709252357483, |
| "learning_rate": 7.356735039120322e-05, |
| "loss": 1.5607, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.31128990755442865, |
| "grad_norm": 0.4739568829536438, |
| "learning_rate": 7.251004440685135e-05, |
| "loss": 1.5571, |
| "step": 3100 |
| }, |
| { |
| "epoch": 0.3213315174755393, |
| "grad_norm": 0.524760365486145, |
| "learning_rate": 7.145273842249948e-05, |
| "loss": 1.5413, |
| "step": 3200 |
| }, |
| { |
| "epoch": 0.33137312739664987, |
| "grad_norm": 0.40299034118652344, |
| "learning_rate": 7.039543243814761e-05, |
| "loss": 1.544, |
| "step": 3300 |
| }, |
| { |
| "epoch": 0.34141473731776045, |
| "grad_norm": 0.6930205225944519, |
| "learning_rate": 6.933812645379572e-05, |
| "loss": 1.5407, |
| "step": 3400 |
| }, |
| { |
| "epoch": 0.3514563472388711, |
| "grad_norm": 0.4104674160480499, |
| "learning_rate": 6.828082046944387e-05, |
| "loss": 1.539, |
| "step": 3500 |
| }, |
| { |
| "epoch": 0.3614979571599817, |
| "grad_norm": 0.4443911910057068, |
| "learning_rate": 6.722351448509198e-05, |
| "loss": 1.534, |
| "step": 3600 |
| }, |
| { |
| "epoch": 0.37153956708109226, |
| "grad_norm": 0.3941650092601776, |
| "learning_rate": 6.616620850074011e-05, |
| "loss": 1.5338, |
| "step": 3700 |
| }, |
| { |
| "epoch": 0.3815811770022029, |
| "grad_norm": 0.530119001865387, |
| "learning_rate": 6.510890251638824e-05, |
| "loss": 1.5244, |
| "step": 3800 |
| }, |
| { |
| "epoch": 0.3916227869233135, |
| "grad_norm": 0.5141203999519348, |
| "learning_rate": 6.405159653203637e-05, |
| "loss": 1.5242, |
| "step": 3900 |
| }, |
| { |
| "epoch": 0.40166439684442407, |
| "grad_norm": 0.5742161870002747, |
| "learning_rate": 6.29942905476845e-05, |
| "loss": 1.5233, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.40166439684442407, |
| "eval_loss": 1.409444808959961, |
| "eval_runtime": 406.7484, |
| "eval_samples_per_second": 174.105, |
| "eval_steps_per_second": 21.765, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.4117060067655347, |
| "grad_norm": 0.49223774671554565, |
| "learning_rate": 6.193698456333263e-05, |
| "loss": 1.5254, |
| "step": 4100 |
| }, |
| { |
| "epoch": 0.4217476166866453, |
| "grad_norm": 0.4922013580799103, |
| "learning_rate": 6.087967857898076e-05, |
| "loss": 1.5187, |
| "step": 4200 |
| }, |
| { |
| "epoch": 0.43178922660775587, |
| "grad_norm": 0.5500440001487732, |
| "learning_rate": 5.982237259462888e-05, |
| "loss": 1.5217, |
| "step": 4300 |
| }, |
| { |
| "epoch": 0.4418308365288665, |
| "grad_norm": 0.5224810242652893, |
| "learning_rate": 5.876506661027702e-05, |
| "loss": 1.5187, |
| "step": 4400 |
| }, |
| { |
| "epoch": 0.4518724464499771, |
| "grad_norm": 0.42916256189346313, |
| "learning_rate": 5.770776062592515e-05, |
| "loss": 1.5064, |
| "step": 4500 |
| }, |
| { |
| "epoch": 0.4619140563710877, |
| "grad_norm": 0.4909166693687439, |
| "learning_rate": 5.665045464157327e-05, |
| "loss": 1.5065, |
| "step": 4600 |
| }, |
| { |
| "epoch": 0.4719556662921983, |
| "grad_norm": 0.4582947790622711, |
| "learning_rate": 5.559314865722141e-05, |
| "loss": 1.5095, |
| "step": 4700 |
| }, |
| { |
| "epoch": 0.4819972762133089, |
| "grad_norm": 0.4197562038898468, |
| "learning_rate": 5.453584267286953e-05, |
| "loss": 1.501, |
| "step": 4800 |
| }, |
| { |
| "epoch": 0.4920388861344195, |
| "grad_norm": 0.42700713872909546, |
| "learning_rate": 5.3478536688517655e-05, |
| "loss": 1.5088, |
| "step": 4900 |
| }, |
| { |
| "epoch": 0.5020804960555301, |
| "grad_norm": 0.6198418140411377, |
| "learning_rate": 5.242123070416579e-05, |
| "loss": 1.4986, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.5121221059766407, |
| "grad_norm": 0.46520060300827026, |
| "learning_rate": 5.1363924719813915e-05, |
| "loss": 1.5118, |
| "step": 5100 |
| }, |
| { |
| "epoch": 0.5221637158977513, |
| "grad_norm": 0.36571305990219116, |
| "learning_rate": 5.030661873546204e-05, |
| "loss": 1.5023, |
| "step": 5200 |
| }, |
| { |
| "epoch": 0.5322053258188619, |
| "grad_norm": 0.3751266300678253, |
| "learning_rate": 4.924931275111017e-05, |
| "loss": 1.4921, |
| "step": 5300 |
| }, |
| { |
| "epoch": 0.5422469357399725, |
| "grad_norm": 0.37080907821655273, |
| "learning_rate": 4.8192006766758305e-05, |
| "loss": 1.492, |
| "step": 5400 |
| }, |
| { |
| "epoch": 0.5522885456610831, |
| "grad_norm": 0.43024304509162903, |
| "learning_rate": 4.7134700782406435e-05, |
| "loss": 1.4859, |
| "step": 5500 |
| }, |
| { |
| "epoch": 0.5623301555821937, |
| "grad_norm": 0.40753471851348877, |
| "learning_rate": 4.607739479805456e-05, |
| "loss": 1.4853, |
| "step": 5600 |
| }, |
| { |
| "epoch": 0.5723717655033043, |
| "grad_norm": 0.40475383400917053, |
| "learning_rate": 4.502008881370269e-05, |
| "loss": 1.489, |
| "step": 5700 |
| }, |
| { |
| "epoch": 0.582413375424415, |
| "grad_norm": 0.3590254485607147, |
| "learning_rate": 4.396278282935082e-05, |
| "loss": 1.4892, |
| "step": 5800 |
| }, |
| { |
| "epoch": 0.5924549853455255, |
| "grad_norm": 0.4830130338668823, |
| "learning_rate": 4.290547684499894e-05, |
| "loss": 1.4922, |
| "step": 5900 |
| }, |
| { |
| "epoch": 0.6024965952666361, |
| "grad_norm": 0.42672935128211975, |
| "learning_rate": 4.184817086064707e-05, |
| "loss": 1.483, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.6024965952666361, |
| "eval_loss": 1.3820070028305054, |
| "eval_runtime": 408.5783, |
| "eval_samples_per_second": 173.325, |
| "eval_steps_per_second": 21.668, |
| "step": 6000 |
| }, |
| { |
| "epoch": 0.6125382051877467, |
| "grad_norm": 0.39759695529937744, |
| "learning_rate": 4.07908648762952e-05, |
| "loss": 1.4778, |
| "step": 6100 |
| }, |
| { |
| "epoch": 0.6225798151088573, |
| "grad_norm": 0.43023234605789185, |
| "learning_rate": 3.9733558891943326e-05, |
| "loss": 1.4774, |
| "step": 6200 |
| }, |
| { |
| "epoch": 0.6326214250299679, |
| "grad_norm": 0.4756367802619934, |
| "learning_rate": 3.867625290759146e-05, |
| "loss": 1.4879, |
| "step": 6300 |
| }, |
| { |
| "epoch": 0.6426630349510786, |
| "grad_norm": 0.46574530005455017, |
| "learning_rate": 3.761894692323959e-05, |
| "loss": 1.4765, |
| "step": 6400 |
| }, |
| { |
| "epoch": 0.6527046448721892, |
| "grad_norm": 0.6158186197280884, |
| "learning_rate": 3.6561640938887716e-05, |
| "loss": 1.4741, |
| "step": 6500 |
| }, |
| { |
| "epoch": 0.6627462547932997, |
| "grad_norm": 0.42495304346084595, |
| "learning_rate": 3.5504334954535846e-05, |
| "loss": 1.4764, |
| "step": 6600 |
| }, |
| { |
| "epoch": 0.6727878647144103, |
| "grad_norm": 0.44197046756744385, |
| "learning_rate": 3.4447028970183976e-05, |
| "loss": 1.4688, |
| "step": 6700 |
| }, |
| { |
| "epoch": 0.6828294746355209, |
| "grad_norm": 0.38747721910476685, |
| "learning_rate": 3.33897229858321e-05, |
| "loss": 1.4772, |
| "step": 6800 |
| }, |
| { |
| "epoch": 0.6928710845566315, |
| "grad_norm": 0.3641980290412903, |
| "learning_rate": 3.233241700148023e-05, |
| "loss": 1.47, |
| "step": 6900 |
| }, |
| { |
| "epoch": 0.7029126944777422, |
| "grad_norm": 0.39210596680641174, |
| "learning_rate": 3.127511101712835e-05, |
| "loss": 1.4734, |
| "step": 7000 |
| }, |
| { |
| "epoch": 0.7129543043988528, |
| "grad_norm": 0.4860726594924927, |
| "learning_rate": 3.0217805032776486e-05, |
| "loss": 1.4815, |
| "step": 7100 |
| }, |
| { |
| "epoch": 0.7229959143199634, |
| "grad_norm": 0.4656274616718292, |
| "learning_rate": 2.9160499048424616e-05, |
| "loss": 1.464, |
| "step": 7200 |
| }, |
| { |
| "epoch": 0.7330375242410739, |
| "grad_norm": 0.37134116888046265, |
| "learning_rate": 2.810319306407274e-05, |
| "loss": 1.4716, |
| "step": 7300 |
| }, |
| { |
| "epoch": 0.7430791341621845, |
| "grad_norm": 0.46362295746803284, |
| "learning_rate": 2.7045887079720873e-05, |
| "loss": 1.4725, |
| "step": 7400 |
| }, |
| { |
| "epoch": 0.7531207440832951, |
| "grad_norm": 0.3737928867340088, |
| "learning_rate": 2.5988581095369003e-05, |
| "loss": 1.4673, |
| "step": 7500 |
| }, |
| { |
| "epoch": 0.7631623540044058, |
| "grad_norm": 0.3332815170288086, |
| "learning_rate": 2.493127511101713e-05, |
| "loss": 1.4608, |
| "step": 7600 |
| }, |
| { |
| "epoch": 0.7732039639255164, |
| "grad_norm": 0.44841036200523376, |
| "learning_rate": 2.3873969126665257e-05, |
| "loss": 1.466, |
| "step": 7700 |
| }, |
| { |
| "epoch": 0.783245573846627, |
| "grad_norm": 0.5019286274909973, |
| "learning_rate": 2.2816663142313387e-05, |
| "loss": 1.4651, |
| "step": 7800 |
| }, |
| { |
| "epoch": 0.7932871837677375, |
| "grad_norm": 0.3360118567943573, |
| "learning_rate": 2.1759357157961517e-05, |
| "loss": 1.4706, |
| "step": 7900 |
| }, |
| { |
| "epoch": 0.8033287936888481, |
| "grad_norm": 0.38790610432624817, |
| "learning_rate": 2.0702051173609644e-05, |
| "loss": 1.4646, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.8033287936888481, |
| "eval_loss": 1.3676984310150146, |
| "eval_runtime": 408.474, |
| "eval_samples_per_second": 173.37, |
| "eval_steps_per_second": 21.673, |
| "step": 8000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 9958, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 2000, |
| "total_flos": 1.35355491680256e+17, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
