| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 0.9999074845036544, |
| "eval_steps": 100, |
| "global_step": 2702, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.001850309926912758, |
| "grad_norm": 0.0, |
| "learning_rate": 3.690036900369004e-07, |
| "loss": 13.4077, |
| "mean_token_accuracy": 0.6696642246642247, |
| "mean_token_accuracy_utility": 0.5283333333333333, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.003700619853825516, |
| "grad_norm": 0.0, |
| "learning_rate": 7.380073800738008e-07, |
| "loss": 13.1868, |
| "mean_token_accuracy": 0.6668213268213267, |
| "mean_token_accuracy_utility": 0.5369841269841269, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.005550929780738274, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1070110701107011e-06, |
| "loss": 13.3219, |
| "mean_token_accuracy": 0.6660182890336829, |
| "mean_token_accuracy_utility": 0.5330976021437832, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.007401239707651032, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4760147601476015e-06, |
| "loss": 13.0294, |
| "mean_token_accuracy": 0.6747476597476599, |
| "mean_token_accuracy_utility": 0.5339621489621489, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.00925154963456379, |
| "grad_norm": 0.0, |
| "learning_rate": 1.845018450184502e-06, |
| "loss": 13.4096, |
| "mean_token_accuracy": 0.6630076789121621, |
| "mean_token_accuracy_utility": 0.5261781038915528, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.011101859561476548, |
| "grad_norm": 0.0, |
| "learning_rate": 2.2140221402214023e-06, |
| "loss": 13.241, |
| "mean_token_accuracy": 0.6639502907336605, |
| "mean_token_accuracy_utility": 0.5340730944232038, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.012952169488389306, |
| "grad_norm": 0.0, |
| "learning_rate": 2.5830258302583027e-06, |
| "loss": 13.0835, |
| "mean_token_accuracy": 0.6655616605616604, |
| "mean_token_accuracy_utility": 0.5328388278388279, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.014802479415302064, |
| "grad_norm": 0.0, |
| "learning_rate": 2.952029520295203e-06, |
| "loss": 12.9643, |
| "mean_token_accuracy": 0.6720201357790516, |
| "mean_token_accuracy_utility": 0.5333986246753721, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.01665278934221482, |
| "grad_norm": 0.0, |
| "learning_rate": 3.3210332103321034e-06, |
| "loss": 13.0751, |
| "mean_token_accuracy": 0.6664652014652014, |
| "mean_token_accuracy_utility": 0.5362332112332112, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.01850309926912758, |
| "grad_norm": 0.0, |
| "learning_rate": 3.690036900369004e-06, |
| "loss": 13.0234, |
| "mean_token_accuracy": 0.6676691560802515, |
| "mean_token_accuracy_utility": 0.5289626241959102, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.020353409196040336, |
| "grad_norm": 0.0, |
| "learning_rate": 4.059040590405905e-06, |
| "loss": 13.057, |
| "mean_token_accuracy": 0.6651879162842768, |
| "mean_token_accuracy_utility": 0.5336956169846984, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.022203719122953096, |
| "grad_norm": 0.0, |
| "learning_rate": 4.428044280442805e-06, |
| "loss": 12.9215, |
| "mean_token_accuracy": 0.6666931871272709, |
| "mean_token_accuracy_utility": 0.528919610221861, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.024054029049865852, |
| "grad_norm": 0.0, |
| "learning_rate": 4.797047970479705e-06, |
| "loss": 12.8981, |
| "mean_token_accuracy": 0.666817256817257, |
| "mean_token_accuracy_utility": 0.5368376068376068, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.025904338976778612, |
| "grad_norm": 0.0, |
| "learning_rate": 5.166051660516605e-06, |
| "loss": 12.7054, |
| "mean_token_accuracy": 0.6826210826210826, |
| "mean_token_accuracy_utility": 0.5329426129426128, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.027754648903691368, |
| "grad_norm": 0.0, |
| "learning_rate": 5.535055350553506e-06, |
| "loss": 12.7254, |
| "mean_token_accuracy": 0.66968253968254, |
| "mean_token_accuracy_utility": 0.5346642246642246, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.029604958830604128, |
| "grad_norm": 0.0, |
| "learning_rate": 5.904059040590406e-06, |
| "loss": 12.6971, |
| "mean_token_accuracy": 0.668555148555149, |
| "mean_token_accuracy_utility": 0.5374236874236874, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.031455268757516884, |
| "grad_norm": 0.0, |
| "learning_rate": 6.273062730627307e-06, |
| "loss": 12.5686, |
| "mean_token_accuracy": 0.6706084656084654, |
| "mean_token_accuracy_utility": 0.5398107448107448, |
| "step": 85 |
| }, |
| { |
| "epoch": 0.03330557868442964, |
| "grad_norm": 0.0, |
| "learning_rate": 6.642066420664207e-06, |
| "loss": 12.3171, |
| "mean_token_accuracy": 0.6794831094831096, |
| "mean_token_accuracy_utility": 0.5412820512820512, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.0351558886113424, |
| "grad_norm": 0.0, |
| "learning_rate": 7.011070110701108e-06, |
| "loss": 12.3366, |
| "mean_token_accuracy": 0.6760337810337812, |
| "mean_token_accuracy_utility": 0.5376739926739926, |
| "step": 95 |
| }, |
| { |
| "epoch": 0.03700619853825516, |
| "grad_norm": 0.0, |
| "learning_rate": 7.380073800738008e-06, |
| "loss": 12.2112, |
| "mean_token_accuracy": 0.6745116540205829, |
| "mean_token_accuracy_utility": 0.5381503466771324, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.03700619853825516, |
| "eval_mean_token_accuracy": 0.7306957748134221, |
| "eval_reasoning_loss": 0.9649235606193542, |
| "eval_reasoning_runtime": 98.0078, |
| "eval_reasoning_samples_per_second": 1.316, |
| "eval_reasoning_steps_per_second": 0.337, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.03700619853825516, |
| "eval_mean_token_accuracy": 0.5379117239242892, |
| "eval_utility_loss": 2.1140193939208984, |
| "eval_utility_runtime": 395.0189, |
| "eval_utility_samples_per_second": 1.321, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.038856508465167916, |
| "grad_norm": 0.0, |
| "learning_rate": 7.749077490774908e-06, |
| "loss": 12.0826, |
| "mean_token_accuracy": 0.6803968253968258, |
| "mean_token_accuracy_utility": 0.539053724053724, |
| "step": 105 |
| }, |
| { |
| "epoch": 0.04070681839208067, |
| "grad_norm": 0.0, |
| "learning_rate": 8.11808118081181e-06, |
| "loss": 12.1166, |
| "mean_token_accuracy": 0.6744770044770044, |
| "mean_token_accuracy_utility": 0.54009768009768, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.04255712831899343, |
| "grad_norm": 0.0, |
| "learning_rate": 8.48708487084871e-06, |
| "loss": 11.8832, |
| "mean_token_accuracy": 0.6795772488918359, |
| "mean_token_accuracy_utility": 0.5404655684093294, |
| "step": 115 |
| }, |
| { |
| "epoch": 0.04440743824590619, |
| "grad_norm": 0.0, |
| "learning_rate": 8.85608856088561e-06, |
| "loss": 11.8737, |
| "mean_token_accuracy": 0.6779161579161578, |
| "mean_token_accuracy_utility": 0.5495970695970696, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.04625774817281895, |
| "grad_norm": 0.0, |
| "learning_rate": 9.22509225092251e-06, |
| "loss": 11.6891, |
| "mean_token_accuracy": 0.6848667615908999, |
| "mean_token_accuracy_utility": 0.5595087096811233, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.048108058099731704, |
| "grad_norm": 0.0, |
| "learning_rate": 9.59409594095941e-06, |
| "loss": 11.3333, |
| "mean_token_accuracy": 0.6931684981684983, |
| "mean_token_accuracy_utility": 0.5651221001221, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.04995836802664446, |
| "grad_norm": 0.0, |
| "learning_rate": 9.963099630996312e-06, |
| "loss": 11.2069, |
| "mean_token_accuracy": 0.6904107524420026, |
| "mean_token_accuracy_utility": 0.5709147970085469, |
| "step": 135 |
| }, |
| { |
| "epoch": 0.051808677953557224, |
| "grad_norm": 0.0, |
| "learning_rate": 1.033210332103321e-05, |
| "loss": 11.226, |
| "mean_token_accuracy": 0.68872216853812, |
| "mean_token_accuracy_utility": 0.5706720001198528, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.05365898788046998, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0701107011070112e-05, |
| "loss": 11.0142, |
| "mean_token_accuracy": 0.6942511192511194, |
| "mean_token_accuracy_utility": 0.5769169719169718, |
| "step": 145 |
| }, |
| { |
| "epoch": 0.055509297807382736, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1070110701107012e-05, |
| "loss": 10.9015, |
| "mean_token_accuracy": 0.6932799526113235, |
| "mean_token_accuracy_utility": 0.5779961459902585, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.05735960773429549, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1439114391143913e-05, |
| "loss": 10.7741, |
| "mean_token_accuracy": 0.6938153676445977, |
| "mean_token_accuracy_utility": 0.5811408526285423, |
| "step": 155 |
| }, |
| { |
| "epoch": 0.059209917661208256, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1808118081180812e-05, |
| "loss": 10.6128, |
| "mean_token_accuracy": 0.6959547878541716, |
| "mean_token_accuracy_utility": 0.5832367689349207, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.06106022758812101, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2177121771217713e-05, |
| "loss": 10.5505, |
| "mean_token_accuracy": 0.6937851037851038, |
| "mean_token_accuracy_utility": 0.5815384615384613, |
| "step": 165 |
| }, |
| { |
| "epoch": 0.06291053751503377, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2546125461254614e-05, |
| "loss": 10.3217, |
| "mean_token_accuracy": 0.6974640007749411, |
| "mean_token_accuracy_utility": 0.5862869962198175, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.06476084744194653, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2915129151291515e-05, |
| "loss": 10.2556, |
| "mean_token_accuracy": 0.6978030801663616, |
| "mean_token_accuracy_utility": 0.5866571037469474, |
| "step": 175 |
| }, |
| { |
| "epoch": 0.06661115736885928, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3284132841328414e-05, |
| "loss": 10.2386, |
| "mean_token_accuracy": 0.7022710622710624, |
| "mean_token_accuracy_utility": 0.5850793650793649, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.06846146729577204, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3653136531365315e-05, |
| "loss": 10.1454, |
| "mean_token_accuracy": 0.7032783882783885, |
| "mean_token_accuracy_utility": 0.5879670329670328, |
| "step": 185 |
| }, |
| { |
| "epoch": 0.0703117772226848, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4022140221402215e-05, |
| "loss": 10.2055, |
| "mean_token_accuracy": 0.7054216358727161, |
| "mean_token_accuracy_utility": 0.5809169222701622, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.07216208714959756, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4391143911439116e-05, |
| "loss": 10.2691, |
| "mean_token_accuracy": 0.6976476397623353, |
| "mean_token_accuracy_utility": 0.5854581818022677, |
| "step": 195 |
| }, |
| { |
| "epoch": 0.07401239707651032, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4760147601476015e-05, |
| "loss": 10.2683, |
| "mean_token_accuracy": 0.6978245828245826, |
| "mean_token_accuracy_utility": 0.5837057387057385, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.07401239707651032, |
| "eval_mean_token_accuracy": 0.7504947122594182, |
| "eval_reasoning_loss": 0.8411272764205933, |
| "eval_reasoning_runtime": 96.9553, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.07401239707651032, |
| "eval_mean_token_accuracy": 0.5844334615587674, |
| "eval_utility_loss": 1.716774582862854, |
| "eval_utility_runtime": 394.8206, |
| "eval_utility_samples_per_second": 1.322, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.07586270700342307, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5129151291512916e-05, |
| "loss": 10.0593, |
| "mean_token_accuracy": 0.7029385429385429, |
| "mean_token_accuracy_utility": 0.5880952380952379, |
| "step": 205 |
| }, |
| { |
| "epoch": 0.07771301693033583, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5498154981549817e-05, |
| "loss": 9.9496, |
| "mean_token_accuracy": 0.7055860805860805, |
| "mean_token_accuracy_utility": 0.5896886446886445, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.0795633268572486, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5867158671586716e-05, |
| "loss": 10.0916, |
| "mean_token_accuracy": 0.7061680911680913, |
| "mean_token_accuracy_utility": 0.5828266178266176, |
| "step": 215 |
| }, |
| { |
| "epoch": 0.08141363678416134, |
| "grad_norm": 0.0, |
| "learning_rate": 1.623616236162362e-05, |
| "loss": 10.2722, |
| "mean_token_accuracy": 0.6947944647944648, |
| "mean_token_accuracy_utility": 0.5856043956043954, |
| "step": 220 |
| }, |
| { |
| "epoch": 0.08326394671107411, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6605166051660518e-05, |
| "loss": 10.0675, |
| "mean_token_accuracy": 0.7027330077330078, |
| "mean_token_accuracy_utility": 0.5866971916971916, |
| "step": 225 |
| }, |
| { |
| "epoch": 0.08511425663798686, |
| "grad_norm": 0.0, |
| "learning_rate": 1.697416974169742e-05, |
| "loss": 9.8428, |
| "mean_token_accuracy": 0.7100119899128587, |
| "mean_token_accuracy_utility": 0.5893888390914447, |
| "step": 230 |
| }, |
| { |
| "epoch": 0.08696456656489962, |
| "grad_norm": 0.0, |
| "learning_rate": 1.734317343173432e-05, |
| "loss": 9.9197, |
| "mean_token_accuracy": 0.7088967951359769, |
| "mean_token_accuracy_utility": 0.5898771985947437, |
| "step": 235 |
| }, |
| { |
| "epoch": 0.08881487649181238, |
| "grad_norm": 0.0, |
| "learning_rate": 1.771217712177122e-05, |
| "loss": 9.7172, |
| "mean_token_accuracy": 0.7104395604395607, |
| "mean_token_accuracy_utility": 0.591135531135531, |
| "step": 240 |
| }, |
| { |
| "epoch": 0.09066518641872513, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8081180811808117e-05, |
| "loss": 9.9688, |
| "mean_token_accuracy": 0.7022303622303622, |
| "mean_token_accuracy_utility": 0.5891697191697189, |
| "step": 245 |
| }, |
| { |
| "epoch": 0.0925154963456379, |
| "grad_norm": 0.0, |
| "learning_rate": 1.845018450184502e-05, |
| "loss": 9.8156, |
| "mean_token_accuracy": 0.7059337099061221, |
| "mean_token_accuracy_utility": 0.5934787853960224, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.09436580627255066, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8819188191881922e-05, |
| "loss": 9.8857, |
| "mean_token_accuracy": 0.7086589336589337, |
| "mean_token_accuracy_utility": 0.5873321123321121, |
| "step": 255 |
| }, |
| { |
| "epoch": 0.09621611619946341, |
| "grad_norm": 0.0, |
| "learning_rate": 1.918819188191882e-05, |
| "loss": 9.7504, |
| "mean_token_accuracy": 0.7102482702482702, |
| "mean_token_accuracy_utility": 0.5934310134310132, |
| "step": 260 |
| }, |
| { |
| "epoch": 0.09806642612637617, |
| "grad_norm": 0.0, |
| "learning_rate": 1.955719557195572e-05, |
| "loss": 9.9074, |
| "mean_token_accuracy": 0.7063057769886334, |
| "mean_token_accuracy_utility": 0.5874277094762788, |
| "step": 265 |
| }, |
| { |
| "epoch": 0.09991673605328892, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9926199261992623e-05, |
| "loss": 10.2155, |
| "mean_token_accuracy": 0.6967224480286374, |
| "mean_token_accuracy_utility": 0.5823895663081337, |
| "step": 270 |
| }, |
| { |
| "epoch": 0.10176704598020168, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9999866396188624e-05, |
| "loss": 9.9839, |
| "mean_token_accuracy": 0.7037715312033531, |
| "mean_token_accuracy_utility": 0.5837810160764818, |
| "step": 275 |
| }, |
| { |
| "epoch": 0.10361735590711445, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9999323636823398e-05, |
| "loss": 9.6569, |
| "mean_token_accuracy": 0.7115974765974767, |
| "mean_token_accuracy_utility": 0.5897252747252746, |
| "step": 280 |
| }, |
| { |
| "epoch": 0.1054676658340272, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9998363394309497e-05, |
| "loss": 9.5531, |
| "mean_token_accuracy": 0.7213462208289794, |
| "mean_token_accuracy_utility": 0.5881290165772922, |
| "step": 285 |
| }, |
| { |
| "epoch": 0.10731797576093996, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9996985708738146e-05, |
| "loss": 9.4367, |
| "mean_token_accuracy": 0.7212759462759465, |
| "mean_token_accuracy_utility": 0.5885286935286934, |
| "step": 290 |
| }, |
| { |
| "epoch": 0.10916828568785271, |
| "grad_norm": 0.0, |
| "learning_rate": 1.999519063762928e-05, |
| "loss": 9.6595, |
| "mean_token_accuracy": 0.7120940170940171, |
| "mean_token_accuracy_utility": 0.5939377289377288, |
| "step": 295 |
| }, |
| { |
| "epoch": 0.11101859561476547, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9992978255929168e-05, |
| "loss": 9.9809, |
| "mean_token_accuracy": 0.7046291882568274, |
| "mean_token_accuracy_utility": 0.5823735232564407, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.11101859561476547, |
| "eval_mean_token_accuracy": 0.7628739452268862, |
| "eval_reasoning_loss": 0.7743347883224487, |
| "eval_reasoning_runtime": 96.9022, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.11101859561476547, |
| "eval_mean_token_accuracy": 0.5883916515891694, |
| "eval_utility_loss": 1.6776387691497803, |
| "eval_utility_runtime": 394.669, |
| "eval_utility_samples_per_second": 1.323, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.11286890554167824, |
| "grad_norm": 0.0, |
| "learning_rate": 1.999034865600726e-05, |
| "loss": 9.9293, |
| "mean_token_accuracy": 0.7012840862840861, |
| "mean_token_accuracy_utility": 0.5918376068376067, |
| "step": 305 |
| }, |
| { |
| "epoch": 0.11471921546859098, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9987301947652354e-05, |
| "loss": 9.8606, |
| "mean_token_accuracy": 0.7106837606837607, |
| "mean_token_accuracy_utility": 0.5828937728937726, |
| "step": 310 |
| }, |
| { |
| "epoch": 0.11656952539550375, |
| "grad_norm": 0.0, |
| "learning_rate": 1.998383825806799e-05, |
| "loss": 9.6, |
| "mean_token_accuracy": 0.7131440781440781, |
| "mean_token_accuracy_utility": 0.5923015873015871, |
| "step": 315 |
| }, |
| { |
| "epoch": 0.11841983532241651, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9979957731867143e-05, |
| "loss": 9.5824, |
| "mean_token_accuracy": 0.7162646607144753, |
| "mean_token_accuracy_utility": 0.5921761555255978, |
| "step": 320 |
| }, |
| { |
| "epoch": 0.12027014524932926, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9975660531066215e-05, |
| "loss": 9.9537, |
| "mean_token_accuracy": 0.7015037081946377, |
| "mean_token_accuracy_utility": 0.5859396960124845, |
| "step": 325 |
| }, |
| { |
| "epoch": 0.12212045517624202, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9970946835078227e-05, |
| "loss": 9.581, |
| "mean_token_accuracy": 0.7173519163076562, |
| "mean_token_accuracy_utility": 0.5931302299974497, |
| "step": 330 |
| }, |
| { |
| "epoch": 0.12397076510315477, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9965816840705355e-05, |
| "loss": 9.5647, |
| "mean_token_accuracy": 0.7198209198209201, |
| "mean_token_accuracy_utility": 0.5895726495726494, |
| "step": 335 |
| }, |
| { |
| "epoch": 0.12582107503006754, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9960270762130705e-05, |
| "loss": 9.6883, |
| "mean_token_accuracy": 0.7125798566190988, |
| "mean_token_accuracy_utility": 0.5877350427350425, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.1276713849569803, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9954308830909372e-05, |
| "loss": 9.5318, |
| "mean_token_accuracy": 0.7186975986975987, |
| "mean_token_accuracy_utility": 0.5909279609279607, |
| "step": 345 |
| }, |
| { |
| "epoch": 0.12952169488389306, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9947931295958778e-05, |
| "loss": 9.4868, |
| "mean_token_accuracy": 0.7168335368335368, |
| "mean_token_accuracy_utility": 0.5940537240537239, |
| "step": 350 |
| }, |
| { |
| "epoch": 0.1313720048108058, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9941138423548266e-05, |
| "loss": 9.5837, |
| "mean_token_accuracy": 0.7140509746628414, |
| "mean_token_accuracy_utility": 0.5909563427919421, |
| "step": 355 |
| }, |
| { |
| "epoch": 0.13322231473771856, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9933930497287996e-05, |
| "loss": 9.6047, |
| "mean_token_accuracy": 0.7142821459051659, |
| "mean_token_accuracy_utility": 0.5865216513907116, |
| "step": 360 |
| }, |
| { |
| "epoch": 0.13507262466463132, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9926307818117098e-05, |
| "loss": 9.6586, |
| "mean_token_accuracy": 0.7144394846700812, |
| "mean_token_accuracy_utility": 0.5858947665865559, |
| "step": 365 |
| }, |
| { |
| "epoch": 0.1369229345915441, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9918270704291104e-05, |
| "loss": 9.5425, |
| "mean_token_accuracy": 0.7178327228327233, |
| "mean_token_accuracy_utility": 0.5872344322344321, |
| "step": 370 |
| }, |
| { |
| "epoch": 0.13877324451845685, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9909819491368677e-05, |
| "loss": 9.2959, |
| "mean_token_accuracy": 0.7283013612923319, |
| "mean_token_accuracy_utility": 0.5946965136694254, |
| "step": 375 |
| }, |
| { |
| "epoch": 0.1406235544453696, |
| "grad_norm": 0.0, |
| "learning_rate": 1.990095453219757e-05, |
| "loss": 9.4032, |
| "mean_token_accuracy": 0.7210927960927963, |
| "mean_token_accuracy_utility": 0.5960989010989008, |
| "step": 380 |
| }, |
| { |
| "epoch": 0.14247386437228235, |
| "grad_norm": 0.0, |
| "learning_rate": 1.989167619689993e-05, |
| "loss": 9.6325, |
| "mean_token_accuracy": 0.7132804232804234, |
| "mean_token_accuracy_utility": 0.5877777777777776, |
| "step": 385 |
| }, |
| { |
| "epoch": 0.1443241742991951, |
| "grad_norm": 0.0, |
| "learning_rate": 1.988198487285682e-05, |
| "loss": 9.7132, |
| "mean_token_accuracy": 0.707126577126577, |
| "mean_token_accuracy_utility": 0.5918681318681317, |
| "step": 390 |
| }, |
| { |
| "epoch": 0.14617448422610788, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9871880964692055e-05, |
| "loss": 9.4251, |
| "mean_token_accuracy": 0.7164143264143262, |
| "mean_token_accuracy_utility": 0.5958608058608057, |
| "step": 395 |
| }, |
| { |
| "epoch": 0.14802479415302064, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9861364894255306e-05, |
| "loss": 9.4035, |
| "mean_token_accuracy": 0.7213918983989406, |
| "mean_token_accuracy_utility": 0.5950059760271026, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.14802479415302064, |
| "eval_mean_token_accuracy": 0.7684223184223187, |
| "eval_reasoning_loss": 0.7462493181228638, |
| "eval_reasoning_runtime": 96.9439, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.14802479415302064, |
| "eval_mean_token_accuracy": 0.590575851740784, |
| "eval_utility_loss": 1.660441517829895, |
| "eval_utility_runtime": 394.8254, |
| "eval_utility_samples_per_second": 1.322, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.1498751040799334, |
| "grad_norm": 0.0, |
| "learning_rate": 1.985043710060449e-05, |
| "loss": 9.7802, |
| "mean_token_accuracy": 0.7107733007733009, |
| "mean_token_accuracy_utility": 0.5856410256410255, |
| "step": 405 |
| }, |
| { |
| "epoch": 0.15172541400684614, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9839098039987435e-05, |
| "loss": 9.4348, |
| "mean_token_accuracy": 0.7218641395220285, |
| "mean_token_accuracy_utility": 0.593614396588063, |
| "step": 410 |
| }, |
| { |
| "epoch": 0.1535757239337589, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9827348185822834e-05, |
| "loss": 9.4574, |
| "mean_token_accuracy": 0.7176794097631636, |
| "mean_token_accuracy_utility": 0.5906816969329581, |
| "step": 415 |
| }, |
| { |
| "epoch": 0.15542603386067166, |
| "grad_norm": 0.0, |
| "learning_rate": 1.981518802868048e-05, |
| "loss": 9.5311, |
| "mean_token_accuracy": 0.7147137168405171, |
| "mean_token_accuracy_utility": 0.5886222250026257, |
| "step": 420 |
| }, |
| { |
| "epoch": 0.15727634378758443, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9802618076260784e-05, |
| "loss": 9.4366, |
| "mean_token_accuracy": 0.7220451770451775, |
| "mean_token_accuracy_utility": 0.5911538461538459, |
| "step": 425 |
| }, |
| { |
| "epoch": 0.1591266537144972, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9789638853373563e-05, |
| "loss": 9.4098, |
| "mean_token_accuracy": 0.7204515410079154, |
| "mean_token_accuracy_utility": 0.5924168940860169, |
| "step": 430 |
| }, |
| { |
| "epoch": 0.16097696364140993, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9776250901916168e-05, |
| "loss": 9.5956, |
| "mean_token_accuracy": 0.7071509971509974, |
| "mean_token_accuracy_utility": 0.5959829059829058, |
| "step": 435 |
| }, |
| { |
| "epoch": 0.1628272735683227, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9762454780850807e-05, |
| "loss": 9.5699, |
| "mean_token_accuracy": 0.7161599511599513, |
| "mean_token_accuracy_utility": 0.5859523809523808, |
| "step": 440 |
| }, |
| { |
| "epoch": 0.16467758349523545, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9748251066181247e-05, |
| "loss": 9.4469, |
| "mean_token_accuracy": 0.720830280830281, |
| "mean_token_accuracy_utility": 0.5896092796092793, |
| "step": 445 |
| }, |
| { |
| "epoch": 0.16652789342214822, |
| "grad_norm": 0.0, |
| "learning_rate": 1.973364035092875e-05, |
| "loss": 9.5715, |
| "mean_token_accuracy": 0.7110600360442112, |
| "mean_token_accuracy_utility": 0.5917759567284812, |
| "step": 450 |
| }, |
| { |
| "epoch": 0.16837820334906098, |
| "grad_norm": 0.0, |
| "learning_rate": 1.971862324510732e-05, |
| "loss": 9.6032, |
| "mean_token_accuracy": 0.7082552015368445, |
| "mean_token_accuracy_utility": 0.592604432449361, |
| "step": 455 |
| }, |
| { |
| "epoch": 0.17022851327597371, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9703200375698223e-05, |
| "loss": 9.5918, |
| "mean_token_accuracy": 0.7112563273164245, |
| "mean_token_accuracy_utility": 0.5927433409236322, |
| "step": 460 |
| }, |
| { |
| "epoch": 0.17207882320288648, |
| "grad_norm": 0.0, |
| "learning_rate": 1.968737238662382e-05, |
| "loss": 9.4302, |
| "mean_token_accuracy": 0.7241677707767729, |
| "mean_token_accuracy_utility": 0.588852518679524, |
| "step": 465 |
| }, |
| { |
| "epoch": 0.17392913312979924, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9671139938720678e-05, |
| "loss": 9.4073, |
| "mean_token_accuracy": 0.720562000357919, |
| "mean_token_accuracy_utility": 0.5888410682288232, |
| "step": 470 |
| }, |
| { |
| "epoch": 0.175779443056712, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9654503709711984e-05, |
| "loss": 9.53, |
| "mean_token_accuracy": 0.7151030543179502, |
| "mean_token_accuracy_utility": 0.5918781495228362, |
| "step": 475 |
| }, |
| { |
| "epoch": 0.17762975298362477, |
| "grad_norm": 0.0, |
| "learning_rate": 1.963746439417924e-05, |
| "loss": 9.2686, |
| "mean_token_accuracy": 0.7258119658119659, |
| "mean_token_accuracy_utility": 0.596166056166056, |
| "step": 480 |
| }, |
| { |
| "epoch": 0.1794800629105375, |
| "grad_norm": 0.0, |
| "learning_rate": 1.962002270353328e-05, |
| "loss": 9.3427, |
| "mean_token_accuracy": 0.7257880766878549, |
| "mean_token_accuracy_utility": 0.5938355365348705, |
| "step": 485 |
| }, |
| { |
| "epoch": 0.18133037283745027, |
| "grad_norm": 0.0, |
| "learning_rate": 1.960217936598454e-05, |
| "loss": 9.3011, |
| "mean_token_accuracy": 0.7263329263329266, |
| "mean_token_accuracy_utility": 0.5915628815628814, |
| "step": 490 |
| }, |
| { |
| "epoch": 0.18318068276436303, |
| "grad_norm": 0.0, |
| "learning_rate": 1.958393512651269e-05, |
| "loss": 9.1344, |
| "mean_token_accuracy": 0.7306213051687469, |
| "mean_token_accuracy_utility": 0.5934878471301713, |
| "step": 495 |
| }, |
| { |
| "epoch": 0.1850309926912758, |
| "grad_norm": 0.0, |
| "learning_rate": 1.956529074683551e-05, |
| "loss": 9.5372, |
| "mean_token_accuracy": 0.7172588522588526, |
| "mean_token_accuracy_utility": 0.5891391941391939, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.1850309926912758, |
| "eval_mean_token_accuracy": 0.7720082749494518, |
| "eval_reasoning_loss": 0.7299458384513855, |
| "eval_reasoning_runtime": 96.983, |
| "eval_reasoning_samples_per_second": 1.33, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.1850309926912758, |
| "eval_mean_token_accuracy": 0.5919271336051658, |
| "eval_utility_loss": 1.6490024328231812, |
| "eval_utility_runtime": 394.9651, |
| "eval_utility_samples_per_second": 1.322, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.18688130261818856, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9546247005377065e-05, |
| "loss": 9.3468, |
| "mean_token_accuracy": 0.7209922562965444, |
| "mean_token_accuracy_utility": 0.5933308592437222, |
| "step": 505 |
| }, |
| { |
| "epoch": 0.18873161254510132, |
| "grad_norm": 0.0, |
| "learning_rate": 1.952680469723526e-05, |
| "loss": 9.3375, |
| "mean_token_accuracy": 0.7243552509617028, |
| "mean_token_accuracy_utility": 0.5978398676592224, |
| "step": 510 |
| }, |
| { |
| "epoch": 0.19058192247201405, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9506964634148597e-05, |
| "loss": 9.4321, |
| "mean_token_accuracy": 0.7180525030525035, |
| "mean_token_accuracy_utility": 0.591141636141636, |
| "step": 515 |
| }, |
| { |
| "epoch": 0.19243223239892682, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9486727644462306e-05, |
| "loss": 9.3861, |
| "mean_token_accuracy": 0.7196491679219551, |
| "mean_token_accuracy_utility": 0.5936117284300895, |
| "step": 520 |
| }, |
| { |
| "epoch": 0.19428254232583958, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9466094573093744e-05, |
| "loss": 9.1721, |
| "mean_token_accuracy": 0.7309291468416305, |
| "mean_token_accuracy_utility": 0.5956812134186642, |
| "step": 525 |
| }, |
| { |
| "epoch": 0.19613285225275234, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9445066281497144e-05, |
| "loss": 9.3084, |
| "mean_token_accuracy": 0.7253794529221659, |
| "mean_token_accuracy_utility": 0.5916206542487925, |
| "step": 530 |
| }, |
| { |
| "epoch": 0.1979831621796651, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9423643647627625e-05, |
| "loss": 9.4506, |
| "mean_token_accuracy": 0.7165669515669516, |
| "mean_token_accuracy_utility": 0.5922527472527471, |
| "step": 535 |
| }, |
| { |
| "epoch": 0.19983347210657784, |
| "grad_norm": 0.0, |
| "learning_rate": 1.940182756590454e-05, |
| "loss": 9.3322, |
| "mean_token_accuracy": 0.7218602589947306, |
| "mean_token_accuracy_utility": 0.5968628282662433, |
| "step": 540 |
| }, |
| { |
| "epoch": 0.2016837820334906, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9379618947174155e-05, |
| "loss": 9.4528, |
| "mean_token_accuracy": 0.7205727390396496, |
| "mean_token_accuracy_utility": 0.5876644930652246, |
| "step": 545 |
| }, |
| { |
| "epoch": 0.20353409196040337, |
| "grad_norm": 0.0, |
| "learning_rate": 1.935701871867158e-05, |
| "loss": 9.4007, |
| "mean_token_accuracy": 0.7220984307654997, |
| "mean_token_accuracy_utility": 0.5888105548117608, |
| "step": 550 |
| }, |
| { |
| "epoch": 0.20538440188731613, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9334027823982103e-05, |
| "loss": 9.2834, |
| "mean_token_accuracy": 0.724535073535074, |
| "mean_token_accuracy_utility": 0.5964407814407813, |
| "step": 555 |
| }, |
| { |
| "epoch": 0.2072347118142289, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9310647223001752e-05, |
| "loss": 9.4702, |
| "mean_token_accuracy": 0.7152360602360602, |
| "mean_token_accuracy_utility": 0.596123321123321, |
| "step": 560 |
| }, |
| { |
| "epoch": 0.20908502174114163, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9286877891897244e-05, |
| "loss": 9.1173, |
| "mean_token_accuracy": 0.7321080885520858, |
| "mean_token_accuracy_utility": 0.5949323242643153, |
| "step": 565 |
| }, |
| { |
| "epoch": 0.2109353316680544, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9262720823065217e-05, |
| "loss": 9.4947, |
| "mean_token_accuracy": 0.711967846967847, |
| "mean_token_accuracy_utility": 0.5918986568986567, |
| "step": 570 |
| }, |
| { |
| "epoch": 0.21278564159496716, |
| "grad_norm": 0.0, |
| "learning_rate": 1.923817702509081e-05, |
| "loss": 9.3774, |
| "mean_token_accuracy": 0.7237912087912088, |
| "mean_token_accuracy_utility": 0.5920207570207568, |
| "step": 575 |
| }, |
| { |
| "epoch": 0.21463595152187992, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9213247522705532e-05, |
| "loss": 9.445, |
| "mean_token_accuracy": 0.7194350880194115, |
| "mean_token_accuracy_utility": 0.5943980601510302, |
| "step": 580 |
| }, |
| { |
| "epoch": 0.21648626144879268, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9187933356744504e-05, |
| "loss": 9.1814, |
| "mean_token_accuracy": 0.726153984482646, |
| "mean_token_accuracy_utility": 0.5964961414821259, |
| "step": 585 |
| }, |
| { |
| "epoch": 0.21833657137570542, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9162235584102973e-05, |
| "loss": 9.2701, |
| "mean_token_accuracy": 0.7217203852737647, |
| "mean_token_accuracy_utility": 0.5958204964806348, |
| "step": 590 |
| }, |
| { |
| "epoch": 0.22018688130261818, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9136155277692215e-05, |
| "loss": 9.2701, |
| "mean_token_accuracy": 0.7235917785917786, |
| "mean_token_accuracy_utility": 0.5913003663003662, |
| "step": 595 |
| }, |
| { |
| "epoch": 0.22203719122953094, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9109693526394722e-05, |
| "loss": 9.232, |
| "mean_token_accuracy": 0.7238074888074885, |
| "mean_token_accuracy_utility": 0.5959401709401708, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.22203719122953094, |
| "eval_mean_token_accuracy": 0.773961659255777, |
| "eval_reasoning_loss": 0.7196130156517029, |
| "eval_reasoning_runtime": 97.0108, |
| "eval_reasoning_samples_per_second": 1.33, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.22203719122953094, |
| "eval_mean_token_accuracy": 0.5929177077458883, |
| "eval_utility_loss": 1.6407129764556885, |
| "eval_utility_runtime": 394.9631, |
| "eval_utility_samples_per_second": 1.322, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.2238875011564437, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9082851435018743e-05, |
| "loss": 9.2364, |
| "mean_token_accuracy": 0.7216877138208354, |
| "mean_token_accuracy_utility": 0.6006553270546908, |
| "step": 605 |
| }, |
| { |
| "epoch": 0.22573781108335647, |
| "grad_norm": 0.0, |
| "learning_rate": 1.905563012425216e-05, |
| "loss": 9.3033, |
| "mean_token_accuracy": 0.7232929540654864, |
| "mean_token_accuracy_utility": 0.5942988866164829, |
| "step": 610 |
| }, |
| { |
| "epoch": 0.2275881210102692, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9028030730615696e-05, |
| "loss": 9.3801, |
| "mean_token_accuracy": 0.7176027676027673, |
| "mean_token_accuracy_utility": 0.5937851037851036, |
| "step": 615 |
| }, |
| { |
| "epoch": 0.22943843093718197, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9000054406415467e-05, |
| "loss": 9.1518, |
| "mean_token_accuracy": 0.7269811888670212, |
| "mean_token_accuracy_utility": 0.5975491832066803, |
| "step": 620 |
| }, |
| { |
| "epoch": 0.23128874086409473, |
| "grad_norm": 0.0, |
| "learning_rate": 1.897170231969486e-05, |
| "loss": 9.1063, |
| "mean_token_accuracy": 0.7245360195360195, |
| "mean_token_accuracy_utility": 0.6014713064713063, |
| "step": 625 |
| }, |
| { |
| "epoch": 0.2331390507910075, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8942975654185788e-05, |
| "loss": 9.2047, |
| "mean_token_accuracy": 0.7261078809882618, |
| "mean_token_accuracy_utility": 0.5978718725130149, |
| "step": 630 |
| }, |
| { |
| "epoch": 0.23498936071792026, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8913875609259246e-05, |
| "loss": 9.1311, |
| "mean_token_accuracy": 0.727435317696946, |
| "mean_token_accuracy_utility": 0.6016588224437059, |
| "step": 635 |
| }, |
| { |
| "epoch": 0.23683967064483302, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8884403399875252e-05, |
| "loss": 9.2464, |
| "mean_token_accuracy": 0.7206244041224827, |
| "mean_token_accuracy_utility": 0.5960978765921124, |
| "step": 640 |
| }, |
| { |
| "epoch": 0.23868998057174576, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8854560256532098e-05, |
| "loss": 9.0903, |
| "mean_token_accuracy": 0.7321853701853699, |
| "mean_token_accuracy_utility": 0.5938949938949938, |
| "step": 645 |
| }, |
| { |
| "epoch": 0.24054029049865852, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8824347425215016e-05, |
| "loss": 9.4335, |
| "mean_token_accuracy": 0.7160003664593507, |
| "mean_token_accuracy_utility": 0.5932025645795176, |
| "step": 650 |
| }, |
| { |
| "epoch": 0.24239060042557128, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8793766167344115e-05, |
| "loss": 9.1711, |
| "mean_token_accuracy": 0.7248692692889123, |
| "mean_token_accuracy_utility": 0.5988813121402405, |
| "step": 655 |
| }, |
| { |
| "epoch": 0.24424091035248405, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8762817759721735e-05, |
| "loss": 9.2323, |
| "mean_token_accuracy": 0.7235999185999185, |
| "mean_token_accuracy_utility": 0.5927045177045176, |
| "step": 660 |
| }, |
| { |
| "epoch": 0.2460912202793968, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8731503494479132e-05, |
| "loss": 9.2042, |
| "mean_token_accuracy": 0.7201798915197609, |
| "mean_token_accuracy_utility": 0.5999291739487816, |
| "step": 665 |
| }, |
| { |
| "epoch": 0.24794153020630955, |
| "grad_norm": 0.0, |
| "learning_rate": 1.869982467902255e-05, |
| "loss": 9.1254, |
| "mean_token_accuracy": 0.7276652532693735, |
| "mean_token_accuracy_utility": 0.5991007659131266, |
| "step": 670 |
| }, |
| { |
| "epoch": 0.2497918401332223, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8667782635978597e-05, |
| "loss": 9.4483, |
| "mean_token_accuracy": 0.7161151811151812, |
| "mean_token_accuracy_utility": 0.5927045177045176, |
| "step": 675 |
| }, |
| { |
| "epoch": 0.2516421500601351, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8635378703139066e-05, |
| "loss": 8.8821, |
| "mean_token_accuracy": 0.7325681725681723, |
| "mean_token_accuracy_utility": 0.6011233211233209, |
| "step": 680 |
| }, |
| { |
| "epoch": 0.25349245998704784, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8602614233405047e-05, |
| "loss": 9.3655, |
| "mean_token_accuracy": 0.723630443630444, |
| "mean_token_accuracy_utility": 0.5910744810744809, |
| "step": 685 |
| }, |
| { |
| "epoch": 0.2553427699139606, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8569490594730474e-05, |
| "loss": 9.2734, |
| "mean_token_accuracy": 0.7217629410970265, |
| "mean_token_accuracy_utility": 0.5947637927660498, |
| "step": 690 |
| }, |
| { |
| "epoch": 0.25719307984087336, |
| "grad_norm": 0.0, |
| "learning_rate": 1.853600917006497e-05, |
| "loss": 9.1792, |
| "mean_token_accuracy": 0.7300405076368058, |
| "mean_token_accuracy_utility": 0.5917210345099287, |
| "step": 695 |
| }, |
| { |
| "epoch": 0.2590433897677861, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8502171357296144e-05, |
| "loss": 8.8668, |
| "mean_token_accuracy": 0.7342409862377279, |
| "mean_token_accuracy_utility": 0.6049451809354053, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.2590433897677861, |
| "eval_mean_token_accuracy": 0.7756934786346551, |
| "eval_reasoning_loss": 0.7123219966888428, |
| "eval_reasoning_runtime": 96.9857, |
| "eval_reasoning_samples_per_second": 1.33, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.2590433897677861, |
| "eval_mean_token_accuracy": 0.5938883888007451, |
| "eval_utility_loss": 1.6342211961746216, |
| "eval_utility_runtime": 395.0229, |
| "eval_utility_samples_per_second": 1.321, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.2608936996946989, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8467978569191216e-05, |
| "loss": 9.3409, |
| "mean_token_accuracy": 0.7232621082621082, |
| "mean_token_accuracy_utility": 0.5961965811965809, |
| "step": 705 |
| }, |
| { |
| "epoch": 0.2627440096216116, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8433432233338027e-05, |
| "loss": 9.0739, |
| "mean_token_accuracy": 0.728923483923484, |
| "mean_token_accuracy_utility": 0.5965262515262514, |
| "step": 710 |
| }, |
| { |
| "epoch": 0.26459431954852436, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8398533792085436e-05, |
| "loss": 9.2825, |
| "mean_token_accuracy": 0.7233182928867271, |
| "mean_token_accuracy_utility": 0.5936667223720242, |
| "step": 715 |
| }, |
| { |
| "epoch": 0.2664446294754371, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8363284702483106e-05, |
| "loss": 9.3506, |
| "mean_token_accuracy": 0.721118922976698, |
| "mean_token_accuracy_utility": 0.5950295406028635, |
| "step": 720 |
| }, |
| { |
| "epoch": 0.2682949394023499, |
| "grad_norm": 0.0, |
| "learning_rate": 1.832768643622067e-05, |
| "loss": 9.0464, |
| "mean_token_accuracy": 0.7311009361009363, |
| "mean_token_accuracy_utility": 0.5984554334554333, |
| "step": 725 |
| }, |
| { |
| "epoch": 0.27014524932926265, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8291740479566286e-05, |
| "loss": 9.1157, |
| "mean_token_accuracy": 0.7278103378103378, |
| "mean_token_accuracy_utility": 0.5980219780219779, |
| "step": 730 |
| }, |
| { |
| "epoch": 0.2719955592561754, |
| "grad_norm": 0.0, |
| "learning_rate": 1.825544833330457e-05, |
| "loss": 9.277, |
| "mean_token_accuracy": 0.7240041923976012, |
| "mean_token_accuracy_utility": 0.593404518584745, |
| "step": 735 |
| }, |
| { |
| "epoch": 0.2738458691830882, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8218811512673958e-05, |
| "loss": 9.3415, |
| "mean_token_accuracy": 0.7189930198309428, |
| "mean_token_accuracy_utility": 0.5981756406894092, |
| "step": 740 |
| }, |
| { |
| "epoch": 0.27569617911000094, |
| "grad_norm": 0.0, |
| "learning_rate": 1.818183154730344e-05, |
| "loss": 9.0994, |
| "mean_token_accuracy": 0.7254354904354904, |
| "mean_token_accuracy_utility": 0.6015567765567764, |
| "step": 745 |
| }, |
| { |
| "epoch": 0.2775464890369137, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8144509981148675e-05, |
| "loss": 9.1737, |
| "mean_token_accuracy": 0.7265214551103728, |
| "mean_token_accuracy_utility": 0.5970368928036452, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.27939679896382646, |
| "grad_norm": 0.0, |
| "learning_rate": 1.810684837242755e-05, |
| "loss": 9.1413, |
| "mean_token_accuracy": 0.7236894586894589, |
| "mean_token_accuracy_utility": 0.6008363858363857, |
| "step": 755 |
| }, |
| { |
| "epoch": 0.2812471088907392, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8068848293555118e-05, |
| "loss": 9.2796, |
| "mean_token_accuracy": 0.7227124515919923, |
| "mean_token_accuracy_utility": 0.5967454133840344, |
| "step": 760 |
| }, |
| { |
| "epoch": 0.28309741881765194, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8030511331077945e-05, |
| "loss": 9.0652, |
| "mean_token_accuracy": 0.7269710121920708, |
| "mean_token_accuracy_utility": 0.6032939889571646, |
| "step": 765 |
| }, |
| { |
| "epoch": 0.2849477287445647, |
| "grad_norm": 0.0, |
| "learning_rate": 1.799183908560787e-05, |
| "loss": 9.4988, |
| "mean_token_accuracy": 0.7159384788428517, |
| "mean_token_accuracy_utility": 0.591747060460178, |
| "step": 770 |
| }, |
| { |
| "epoch": 0.28679803867147746, |
| "grad_norm": 0.0, |
| "learning_rate": 1.795283317175518e-05, |
| "loss": 9.0361, |
| "mean_token_accuracy": 0.7305634850413653, |
| "mean_token_accuracy_utility": 0.5994316028652438, |
| "step": 775 |
| }, |
| { |
| "epoch": 0.2886483485983902, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7913495218061202e-05, |
| "loss": 9.1516, |
| "mean_token_accuracy": 0.7254392961981494, |
| "mean_token_accuracy_utility": 0.5970871193636791, |
| "step": 780 |
| }, |
| { |
| "epoch": 0.290498658525303, |
| "grad_norm": 0.0, |
| "learning_rate": 1.787382686693029e-05, |
| "loss": 9.1817, |
| "mean_token_accuracy": 0.7255677655677657, |
| "mean_token_accuracy_utility": 0.5926129426129425, |
| "step": 785 |
| }, |
| { |
| "epoch": 0.29234896845221575, |
| "grad_norm": 0.0, |
| "learning_rate": 1.783382977456128e-05, |
| "loss": 9.1677, |
| "mean_token_accuracy": 0.728431960086854, |
| "mean_token_accuracy_utility": 0.5946487496134315, |
| "step": 790 |
| }, |
| { |
| "epoch": 0.2941992783791285, |
| "grad_norm": 0.0, |
| "learning_rate": 1.779350561087833e-05, |
| "loss": 9.434, |
| "mean_token_accuracy": 0.7189491735616264, |
| "mean_token_accuracy_utility": 0.5930990469364038, |
| "step": 795 |
| }, |
| { |
| "epoch": 0.2960495883060413, |
| "grad_norm": 0.0, |
| "learning_rate": 1.775285605946119e-05, |
| "loss": 9.2522, |
| "mean_token_accuracy": 0.7282112332112336, |
| "mean_token_accuracy_utility": 0.5889194139194137, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.2960495883060413, |
| "eval_mean_token_accuracy": 0.7766848619789801, |
| "eval_reasoning_loss": 0.7065660357475281, |
| "eval_reasoning_runtime": 97.0182, |
| "eval_reasoning_samples_per_second": 1.33, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.2960495883060413, |
| "eval_mean_token_accuracy": 0.5944469875670363, |
| "eval_utility_loss": 1.6288857460021973, |
| "eval_utility_runtime": 395.1756, |
| "eval_utility_samples_per_second": 1.321, |
| "eval_utility_steps_per_second": 0.331, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.29789989823295404, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7711882817474922e-05, |
| "loss": 8.9582, |
| "mean_token_accuracy": 0.730512820512821, |
| "mean_token_accuracy_utility": 0.6019780219780217, |
| "step": 805 |
| }, |
| { |
| "epoch": 0.2997502081598668, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7670587595599034e-05, |
| "loss": 8.9867, |
| "mean_token_accuracy": 0.7276490435954519, |
| "mean_token_accuracy_utility": 0.5998579976972221, |
| "step": 810 |
| }, |
| { |
| "epoch": 0.3016005180867795, |
| "grad_norm": 0.0, |
| "learning_rate": 1.762897211795607e-05, |
| "loss": 9.264, |
| "mean_token_accuracy": 0.7203622303622301, |
| "mean_token_accuracy_utility": 0.5988278388278386, |
| "step": 815 |
| }, |
| { |
| "epoch": 0.3034508280136923, |
| "grad_norm": 0.0, |
| "learning_rate": 1.758703812203961e-05, |
| "loss": 9.2448, |
| "mean_token_accuracy": 0.7203968253968255, |
| "mean_token_accuracy_utility": 0.6027533577533576, |
| "step": 820 |
| }, |
| { |
| "epoch": 0.30530113794060504, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7544787358641735e-05, |
| "loss": 9.4283, |
| "mean_token_accuracy": 0.7131908831908831, |
| "mean_token_accuracy_utility": 0.5980586080586078, |
| "step": 825 |
| }, |
| { |
| "epoch": 0.3071514478675178, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7502221591779932e-05, |
| "loss": 9.3465, |
| "mean_token_accuracy": 0.7137769637769639, |
| "mean_token_accuracy_utility": 0.5996336996336995, |
| "step": 830 |
| }, |
| { |
| "epoch": 0.30900175779443056, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7459342598623438e-05, |
| "loss": 9.1292, |
| "mean_token_accuracy": 0.7275885225885229, |
| "mean_token_accuracy_utility": 0.5996398046398045, |
| "step": 835 |
| }, |
| { |
| "epoch": 0.31085206772134333, |
| "grad_norm": 0.0, |
| "learning_rate": 1.741615216941905e-05, |
| "loss": 9.0636, |
| "mean_token_accuracy": 0.7322323972323976, |
| "mean_token_accuracy_utility": 0.5954761904761903, |
| "step": 840 |
| }, |
| { |
| "epoch": 0.3127023776482561, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7372652107416364e-05, |
| "loss": 9.1874, |
| "mean_token_accuracy": 0.7233967868019595, |
| "mean_token_accuracy_utility": 0.5993844996000166, |
| "step": 845 |
| }, |
| { |
| "epoch": 0.31455268757516885, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7328844228792513e-05, |
| "loss": 8.9323, |
| "mean_token_accuracy": 0.7374989824989825, |
| "mean_token_accuracy_utility": 0.5994322344322343, |
| "step": 850 |
| }, |
| { |
| "epoch": 0.3164029975020816, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7284730362576308e-05, |
| "loss": 9.1077, |
| "mean_token_accuracy": 0.7258343508343509, |
| "mean_token_accuracy_utility": 0.6013125763125762, |
| "step": 855 |
| }, |
| { |
| "epoch": 0.3182533074289944, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7240312350571905e-05, |
| "loss": 9.1825, |
| "mean_token_accuracy": 0.7270253076629855, |
| "mean_token_accuracy_utility": 0.5937987557117889, |
| "step": 860 |
| }, |
| { |
| "epoch": 0.3201036173559071, |
| "grad_norm": 0.0, |
| "learning_rate": 1.719559204728188e-05, |
| "loss": 9.1025, |
| "mean_token_accuracy": 0.7316198616198618, |
| "mean_token_accuracy_utility": 0.595128205128205, |
| "step": 865 |
| }, |
| { |
| "epoch": 0.32195392728281985, |
| "grad_norm": 0.0, |
| "learning_rate": 1.715057131982983e-05, |
| "loss": 9.1669, |
| "mean_token_accuracy": 0.7263469671117204, |
| "mean_token_accuracy_utility": 0.5975390698333289, |
| "step": 870 |
| }, |
| { |
| "epoch": 0.3238042372097326, |
| "grad_norm": 0.0, |
| "learning_rate": 1.710525204788239e-05, |
| "loss": 8.9722, |
| "mean_token_accuracy": 0.7309566079564707, |
| "mean_token_accuracy_utility": 0.5979247689243565, |
| "step": 875 |
| }, |
| { |
| "epoch": 0.3256545471366454, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7059636123570767e-05, |
| "loss": 9.1151, |
| "mean_token_accuracy": 0.7268384872925281, |
| "mean_token_accuracy_utility": 0.59676698812911, |
| "step": 880 |
| }, |
| { |
| "epoch": 0.32750485706355814, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7013725451411757e-05, |
| "loss": 9.0295, |
| "mean_token_accuracy": 0.7344591783247647, |
| "mean_token_accuracy_utility": 0.5961003676971264, |
| "step": 885 |
| }, |
| { |
| "epoch": 0.3293551669904709, |
| "grad_norm": 0.0, |
| "learning_rate": 1.696752194822819e-05, |
| "loss": 9.0816, |
| "mean_token_accuracy": 0.7328681222159485, |
| "mean_token_accuracy_utility": 0.5936104716539498, |
| "step": 890 |
| }, |
| { |
| "epoch": 0.33120547691738367, |
| "grad_norm": 0.0, |
| "learning_rate": 1.692102754306895e-05, |
| "loss": 9.0356, |
| "mean_token_accuracy": 0.7323439534528149, |
| "mean_token_accuracy_utility": 0.5960184293450126, |
| "step": 895 |
| }, |
| { |
| "epoch": 0.33305578684429643, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6874244177128395e-05, |
| "loss": 9.035, |
| "mean_token_accuracy": 0.7324623524623527, |
| "mean_token_accuracy_utility": 0.598192918192918, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.33305578684429643, |
| "eval_mean_token_accuracy": 0.7779025984908339, |
| "eval_reasoning_loss": 0.7023798227310181, |
| "eval_reasoning_runtime": 97.0444, |
| "eval_reasoning_samples_per_second": 1.329, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.33305578684429643, |
| "eval_mean_token_accuracy": 0.5950219329960279, |
| "eval_utility_loss": 1.6242883205413818, |
| "eval_utility_runtime": 393.3522, |
| "eval_utility_samples_per_second": 1.327, |
| "eval_utility_steps_per_second": 0.333, |
| "step": 900 |
| }, |
| { |
| "epoch": 0.3349060967712092, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6827173803665328e-05, |
| "loss": 9.4144, |
| "mean_token_accuracy": 0.7155367501039326, |
| "mean_token_accuracy_utility": 0.5924466361481835, |
| "step": 905 |
| }, |
| { |
| "epoch": 0.33675640669812196, |
| "grad_norm": 0.0, |
| "learning_rate": 1.677981838792144e-05, |
| "loss": 9.1262, |
| "mean_token_accuracy": 0.7248229548229549, |
| "mean_token_accuracy_utility": 0.6013919413919412, |
| "step": 910 |
| }, |
| { |
| "epoch": 0.3386067166250347, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6732179907039266e-05, |
| "loss": 9.3463, |
| "mean_token_accuracy": 0.7240476190476192, |
| "mean_token_accuracy_utility": 0.590164835164835, |
| "step": 915 |
| }, |
| { |
| "epoch": 0.34045702655194743, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6684260349979637e-05, |
| "loss": 9.0944, |
| "mean_token_accuracy": 0.7258047388924014, |
| "mean_token_accuracy_utility": 0.601675510938498, |
| "step": 920 |
| }, |
| { |
| "epoch": 0.3423073364788602, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6636061717438626e-05, |
| "loss": 8.9222, |
| "mean_token_accuracy": 0.7352157102157102, |
| "mean_token_accuracy_utility": 0.6023504273504272, |
| "step": 925 |
| }, |
| { |
| "epoch": 0.34415764640577295, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6587586021764022e-05, |
| "loss": 9.0598, |
| "mean_token_accuracy": 0.7309198209198208, |
| "mean_token_accuracy_utility": 0.5979609279609277, |
| "step": 930 |
| }, |
| { |
| "epoch": 0.3460079563326857, |
| "grad_norm": 0.0, |
| "learning_rate": 1.653883528687133e-05, |
| "loss": 9.2388, |
| "mean_token_accuracy": 0.7249570086795061, |
| "mean_token_accuracy_utility": 0.5940904728218159, |
| "step": 935 |
| }, |
| { |
| "epoch": 0.3478582662595985, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6489811548159245e-05, |
| "loss": 9.0249, |
| "mean_token_accuracy": 0.7237423687423689, |
| "mean_token_accuracy_utility": 0.6017032967032965, |
| "step": 940 |
| }, |
| { |
| "epoch": 0.34970857618651124, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6440516852424678e-05, |
| "loss": 9.1896, |
| "mean_token_accuracy": 0.7223641104930525, |
| "mean_token_accuracy_utility": 0.5980080823949078, |
| "step": 945 |
| }, |
| { |
| "epoch": 0.351558886113424, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6390953257777324e-05, |
| "loss": 9.1597, |
| "mean_token_accuracy": 0.7292912359003183, |
| "mean_token_accuracy_utility": 0.59313622296347, |
| "step": 950 |
| }, |
| { |
| "epoch": 0.35340919604033677, |
| "grad_norm": 0.0, |
| "learning_rate": 1.634112283355369e-05, |
| "loss": 9.0807, |
| "mean_token_accuracy": 0.7234892311677033, |
| "mean_token_accuracy_utility": 0.5995885726239887, |
| "step": 955 |
| }, |
| { |
| "epoch": 0.35525950596724953, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6291027660230735e-05, |
| "loss": 9.0863, |
| "mean_token_accuracy": 0.7282498982498985, |
| "mean_token_accuracy_utility": 0.5965079365079362, |
| "step": 960 |
| }, |
| { |
| "epoch": 0.3571098158941623, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6240669829338992e-05, |
| "loss": 9.1927, |
| "mean_token_accuracy": 0.7197720797720797, |
| "mean_token_accuracy_utility": 0.6005860805860804, |
| "step": 965 |
| }, |
| { |
| "epoch": 0.358960125821075, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6190051443375248e-05, |
| "loss": 9.0902, |
| "mean_token_accuracy": 0.7283068783068785, |
| "mean_token_accuracy_utility": 0.6005372405372403, |
| "step": 970 |
| }, |
| { |
| "epoch": 0.36081043574798777, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6139174615714753e-05, |
| "loss": 9.1797, |
| "mean_token_accuracy": 0.7259492817632355, |
| "mean_token_accuracy_utility": 0.594245891687752, |
| "step": 975 |
| }, |
| { |
| "epoch": 0.36266074567490053, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6088041470523005e-05, |
| "loss": 9.2008, |
| "mean_token_accuracy": 0.7200020350020352, |
| "mean_token_accuracy_utility": 0.5986263736263735, |
| "step": 980 |
| }, |
| { |
| "epoch": 0.3645110556018133, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6036654142667043e-05, |
| "loss": 9.2383, |
| "mean_token_accuracy": 0.7227501766894822, |
| "mean_token_accuracy_utility": 0.598861030678947, |
| "step": 985 |
| }, |
| { |
| "epoch": 0.36636136552872606, |
| "grad_norm": 0.0, |
| "learning_rate": 1.598501477762632e-05, |
| "loss": 9.1678, |
| "mean_token_accuracy": 0.7235185185185187, |
| "mean_token_accuracy_utility": 0.5986874236874236, |
| "step": 990 |
| }, |
| { |
| "epoch": 0.3682116754556388, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5933125531403135e-05, |
| "loss": 9.1243, |
| "mean_token_accuracy": 0.7241715986997422, |
| "mean_token_accuracy_utility": 0.5966170674696362, |
| "step": 995 |
| }, |
| { |
| "epoch": 0.3700619853825516, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5880988570432603e-05, |
| "loss": 8.9797, |
| "mean_token_accuracy": 0.7330635031370439, |
| "mean_token_accuracy_utility": 0.5952100454306681, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.3700619853825516, |
| "eval_mean_token_accuracy": 0.7789258671611614, |
| "eval_reasoning_loss": 0.6987125873565674, |
| "eval_reasoning_runtime": 97.0871, |
| "eval_reasoning_samples_per_second": 1.329, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.3700619853825516, |
| "eval_mean_token_accuracy": 0.5955394760720144, |
| "eval_utility_loss": 1.620626449584961, |
| "eval_utility_runtime": 400.0157, |
| "eval_utility_samples_per_second": 1.305, |
| "eval_utility_steps_per_second": 0.327, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.37191229530946435, |
| "grad_norm": 0.0, |
| "learning_rate": 1.582860607149222e-05, |
| "loss": 9.0893, |
| "mean_token_accuracy": 0.7264387464387467, |
| "mean_token_accuracy_utility": 0.5978876678876677, |
| "step": 1005 |
| }, |
| { |
| "epoch": 0.3737626052363771, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5775980221610966e-05, |
| "loss": 9.1683, |
| "mean_token_accuracy": 0.7241473341473341, |
| "mean_token_accuracy_utility": 0.5982783882783881, |
| "step": 1010 |
| }, |
| { |
| "epoch": 0.3756129151632899, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5723113217978e-05, |
| "loss": 9.0588, |
| "mean_token_accuracy": 0.7279181929181928, |
| "mean_token_accuracy_utility": 0.6030219780219779, |
| "step": 1015 |
| }, |
| { |
| "epoch": 0.37746322509020264, |
| "grad_norm": 0.0, |
| "learning_rate": 1.567000726785093e-05, |
| "loss": 8.898, |
| "mean_token_accuracy": 0.7335917785917789, |
| "mean_token_accuracy_utility": 0.5977350427350425, |
| "step": 1020 |
| }, |
| { |
| "epoch": 0.37931353501711534, |
| "grad_norm": 0.0, |
| "learning_rate": 1.561666458846365e-05, |
| "loss": 9.3211, |
| "mean_token_accuracy": 0.717739112739113, |
| "mean_token_accuracy_utility": 0.5948778998778997, |
| "step": 1025 |
| }, |
| { |
| "epoch": 0.3811638449440281, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5563087406933762e-05, |
| "loss": 9.2303, |
| "mean_token_accuracy": 0.7223109208135564, |
| "mean_token_accuracy_utility": 0.5997776952856012, |
| "step": 1030 |
| }, |
| { |
| "epoch": 0.38301415487094087, |
| "grad_norm": 0.0, |
| "learning_rate": 1.550927796016961e-05, |
| "loss": 9.0998, |
| "mean_token_accuracy": 0.729191659035409, |
| "mean_token_accuracy_utility": 0.5995774191086689, |
| "step": 1035 |
| }, |
| { |
| "epoch": 0.38486446479785363, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5455238494776876e-05, |
| "loss": 9.2099, |
| "mean_token_accuracy": 0.7216671358375125, |
| "mean_token_accuracy_utility": 0.5982614807726111, |
| "step": 1040 |
| }, |
| { |
| "epoch": 0.3867147747247664, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5400971266964772e-05, |
| "loss": 9.0856, |
| "mean_token_accuracy": 0.7319063644688648, |
| "mean_token_accuracy_utility": 0.5940829517704516, |
| "step": 1045 |
| }, |
| { |
| "epoch": 0.38856508465167916, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5346478542451862e-05, |
| "loss": 9.1017, |
| "mean_token_accuracy": 0.7250884937418444, |
| "mean_token_accuracy_utility": 0.6009675569276084, |
| "step": 1050 |
| }, |
| { |
| "epoch": 0.3904153945785919, |
| "grad_norm": 0.0, |
| "learning_rate": 1.529176259637145e-05, |
| "loss": 9.1244, |
| "mean_token_accuracy": 0.7273052464836506, |
| "mean_token_accuracy_utility": 0.6037228212580323, |
| "step": 1055 |
| }, |
| { |
| "epoch": 0.3922657045055047, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5236825713176584e-05, |
| "loss": 9.0907, |
| "mean_token_accuracy": 0.7234798534798536, |
| "mean_token_accuracy_utility": 0.6017948717948716, |
| "step": 1060 |
| }, |
| { |
| "epoch": 0.39411601443241745, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5181670186544706e-05, |
| "loss": 9.1343, |
| "mean_token_accuracy": 0.7219617419617422, |
| "mean_token_accuracy_utility": 0.5995115995115994, |
| "step": 1065 |
| }, |
| { |
| "epoch": 0.3959663243593302, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5126298319281859e-05, |
| "loss": 9.048, |
| "mean_token_accuracy": 0.733862433862434, |
| "mean_token_accuracy_utility": 0.5950061050061048, |
| "step": 1070 |
| }, |
| { |
| "epoch": 0.3978166342862429, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5070712423226552e-05, |
| "loss": 8.9767, |
| "mean_token_accuracy": 0.731427139925057, |
| "mean_token_accuracy_utility": 0.5999346554284062, |
| "step": 1075 |
| }, |
| { |
| "epoch": 0.3996669442131557, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5014914819153252e-05, |
| "loss": 9.2216, |
| "mean_token_accuracy": 0.720816035816036, |
| "mean_token_accuracy_utility": 0.5976129426129424, |
| "step": 1080 |
| }, |
| { |
| "epoch": 0.40151725414006845, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4958907836675467e-05, |
| "loss": 9.2034, |
| "mean_token_accuracy": 0.7247639397639397, |
| "mean_token_accuracy_utility": 0.597197802197802, |
| "step": 1085 |
| }, |
| { |
| "epoch": 0.4033675640669812, |
| "grad_norm": 0.0, |
| "learning_rate": 1.490269381414849e-05, |
| "loss": 9.0578, |
| "mean_token_accuracy": 0.7261050061050062, |
| "mean_token_accuracy_utility": 0.597179487179487, |
| "step": 1090 |
| }, |
| { |
| "epoch": 0.405217873993894, |
| "grad_norm": 0.0, |
| "learning_rate": 1.484627509857178e-05, |
| "loss": 9.0941, |
| "mean_token_accuracy": 0.728356812489307, |
| "mean_token_accuracy_utility": 0.5984159808134647, |
| "step": 1095 |
| }, |
| { |
| "epoch": 0.40706818392080674, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4789654045490957e-05, |
| "loss": 8.985, |
| "mean_token_accuracy": 0.7313247863247867, |
| "mean_token_accuracy_utility": 0.5983699633699631, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.40706818392080674, |
| "eval_mean_token_accuracy": 0.7792274174627113, |
| "eval_reasoning_loss": 0.6963344812393188, |
| "eval_reasoning_runtime": 96.286, |
| "eval_reasoning_samples_per_second": 1.34, |
| "eval_reasoning_steps_per_second": 0.343, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.40706818392080674, |
| "eval_mean_token_accuracy": 0.5958862170775925, |
| "eval_utility_loss": 1.6175833940505981, |
| "eval_utility_runtime": 391.9587, |
| "eval_utility_samples_per_second": 1.332, |
| "eval_utility_steps_per_second": 0.334, |
| "step": 1100 |
| }, |
| { |
| "epoch": 0.4089184938477195, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4732833018899468e-05, |
| "loss": 9.234, |
| "mean_token_accuracy": 0.7215873015873018, |
| "mean_token_accuracy_utility": 0.5969963369963368, |
| "step": 1105 |
| }, |
| { |
| "epoch": 0.41076880377463226, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4675814391139875e-05, |
| "loss": 9.0476, |
| "mean_token_accuracy": 0.7258682663967921, |
| "mean_token_accuracy_utility": 0.6029894145749907, |
| "step": 1110 |
| }, |
| { |
| "epoch": 0.412619113701545, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4618600542804819e-05, |
| "loss": 9.5661, |
| "mean_token_accuracy": 0.7116666666666668, |
| "mean_token_accuracy_utility": 0.5920207570207567, |
| "step": 1115 |
| }, |
| { |
| "epoch": 0.4144694236284578, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4561193862637621e-05, |
| "loss": 9.0448, |
| "mean_token_accuracy": 0.7286223036223038, |
| "mean_token_accuracy_utility": 0.6007020757020756, |
| "step": 1120 |
| }, |
| { |
| "epoch": 0.4163197335553705, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4503596747432554e-05, |
| "loss": 9.3536, |
| "mean_token_accuracy": 0.7228035124768889, |
| "mean_token_accuracy_utility": 0.5854679244880535, |
| "step": 1125 |
| }, |
| { |
| "epoch": 0.41817004348228326, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4445811601934763e-05, |
| "loss": 9.0765, |
| "mean_token_accuracy": 0.7284126984126986, |
| "mean_token_accuracy_utility": 0.5957997557997555, |
| "step": 1130 |
| }, |
| { |
| "epoch": 0.420020353409196, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4387840838739875e-05, |
| "loss": 9.357, |
| "mean_token_accuracy": 0.7202604802604803, |
| "mean_token_accuracy_utility": 0.5926251526251525, |
| "step": 1135 |
| }, |
| { |
| "epoch": 0.4218706633361088, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4329686878193271e-05, |
| "loss": 9.1472, |
| "mean_token_accuracy": 0.7225722425722424, |
| "mean_token_accuracy_utility": 0.5973260073260072, |
| "step": 1140 |
| }, |
| { |
| "epoch": 0.42372097326302155, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4271352148289025e-05, |
| "loss": 8.9256, |
| "mean_token_accuracy": 0.735403301439571, |
| "mean_token_accuracy_utility": 0.599799647908456, |
| "step": 1145 |
| }, |
| { |
| "epoch": 0.4255712831899343, |
| "grad_norm": 0.0, |
| "learning_rate": 1.421283908456854e-05, |
| "loss": 8.8196, |
| "mean_token_accuracy": 0.7369658119658117, |
| "mean_token_accuracy_utility": 0.601202686202686, |
| "step": 1150 |
| }, |
| { |
| "epoch": 0.4274215931168471, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4154150130018867e-05, |
| "loss": 9.1622, |
| "mean_token_accuracy": 0.7307834757834759, |
| "mean_token_accuracy_utility": 0.591288156288156, |
| "step": 1155 |
| }, |
| { |
| "epoch": 0.42927190304375984, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4095287734970678e-05, |
| "loss": 9.0175, |
| "mean_token_accuracy": 0.7295318515481457, |
| "mean_token_accuracy_utility": 0.5998776059264885, |
| "step": 1160 |
| }, |
| { |
| "epoch": 0.4311222129706726, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4036254356996004e-05, |
| "loss": 9.1139, |
| "mean_token_accuracy": 0.7225824175824176, |
| "mean_token_accuracy_utility": 0.5985409035409034, |
| "step": 1165 |
| }, |
| { |
| "epoch": 0.43297252289758537, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3977052460805597e-05, |
| "loss": 9.1498, |
| "mean_token_accuracy": 0.7288665038665039, |
| "mean_token_accuracy_utility": 0.5947313797313796, |
| "step": 1170 |
| }, |
| { |
| "epoch": 0.43482283282449813, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3917684518146044e-05, |
| "loss": 9.1134, |
| "mean_token_accuracy": 0.7275824175824176, |
| "mean_token_accuracy_utility": 0.5985714285714284, |
| "step": 1175 |
| }, |
| { |
| "epoch": 0.43667314275141084, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3858153007696552e-05, |
| "loss": 8.9865, |
| "mean_token_accuracy": 0.7275295075295076, |
| "mean_token_accuracy_utility": 0.6020634920634919, |
| "step": 1180 |
| }, |
| { |
| "epoch": 0.4385234526783236, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3798460414965475e-05, |
| "loss": 8.9922, |
| "mean_token_accuracy": 0.7334676434676434, |
| "mean_token_accuracy_utility": 0.5966178266178265, |
| "step": 1185 |
| }, |
| { |
| "epoch": 0.44037376260523636, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3738609232186537e-05, |
| "loss": 9.3115, |
| "mean_token_accuracy": 0.7185510785510786, |
| "mean_token_accuracy_utility": 0.5961050061050058, |
| "step": 1190 |
| }, |
| { |
| "epoch": 0.4422240725321491, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3678601958214779e-05, |
| "loss": 9.1859, |
| "mean_token_accuracy": 0.7306878306878308, |
| "mean_token_accuracy_utility": 0.5893406593406592, |
| "step": 1195 |
| }, |
| { |
| "epoch": 0.4440743824590619, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3618441098422215e-05, |
| "loss": 8.9813, |
| "mean_token_accuracy": 0.7343182743182743, |
| "mean_token_accuracy_utility": 0.5960317460317459, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.4440743824590619, |
| "eval_mean_token_accuracy": 0.7796860002742354, |
| "eval_reasoning_loss": 0.6942024827003479, |
| "eval_reasoning_runtime": 97.0085, |
| "eval_reasoning_samples_per_second": 1.33, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.4440743824590619, |
| "eval_mean_token_accuracy": 0.596267911205667, |
| "eval_utility_loss": 1.6151468753814697, |
| "eval_utility_runtime": 394.9976, |
| "eval_utility_samples_per_second": 1.322, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 1200 |
| }, |
| { |
| "epoch": 0.44592469238597465, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3558129164593256e-05, |
| "loss": 9.1919, |
| "mean_token_accuracy": 0.7252766141984119, |
| "mean_token_accuracy_utility": 0.5985526753180677, |
| "step": 1205 |
| }, |
| { |
| "epoch": 0.4477750023128874, |
| "grad_norm": 0.0, |
| "learning_rate": 1.349766867481982e-05, |
| "loss": 9.0929, |
| "mean_token_accuracy": 0.7245787545787543, |
| "mean_token_accuracy_utility": 0.6009768009768008, |
| "step": 1210 |
| }, |
| { |
| "epoch": 0.4496253122398002, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3437062153396201e-05, |
| "loss": 9.189, |
| "mean_token_accuracy": 0.7205352055352058, |
| "mean_token_accuracy_utility": 0.5994444444444442, |
| "step": 1215 |
| }, |
| { |
| "epoch": 0.45147562216671294, |
| "grad_norm": 0.0, |
| "learning_rate": 1.337631213071369e-05, |
| "loss": 9.0773, |
| "mean_token_accuracy": 0.7220255418284094, |
| "mean_token_accuracy_utility": 0.6008141102227122, |
| "step": 1220 |
| }, |
| { |
| "epoch": 0.4533259320936257, |
| "grad_norm": 0.0, |
| "learning_rate": 1.331542114315491e-05, |
| "loss": 9.0862, |
| "mean_token_accuracy": 0.7286389742432398, |
| "mean_token_accuracy_utility": 0.5957826125954087, |
| "step": 1225 |
| }, |
| { |
| "epoch": 0.4551762420205384, |
| "grad_norm": 0.0, |
| "learning_rate": 1.325439173298793e-05, |
| "loss": 8.8788, |
| "mean_token_accuracy": 0.7355923901413589, |
| "mean_token_accuracy_utility": 0.6005012241481312, |
| "step": 1230 |
| }, |
| { |
| "epoch": 0.4570265519474512, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3193226448260128e-05, |
| "loss": 9.1452, |
| "mean_token_accuracy": 0.7243060833323417, |
| "mean_token_accuracy_utility": 0.5986813757601501, |
| "step": 1235 |
| }, |
| { |
| "epoch": 0.45887686187436394, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3131927842691793e-05, |
| "loss": 9.3095, |
| "mean_token_accuracy": 0.7212222599714434, |
| "mean_token_accuracy_utility": 0.5923847286322791, |
| "step": 1240 |
| }, |
| { |
| "epoch": 0.4607271718012767, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3070498475569507e-05, |
| "loss": 9.2738, |
| "mean_token_accuracy": 0.7150737893967742, |
| "mean_token_accuracy_utility": 0.6004350434039971, |
| "step": 1245 |
| }, |
| { |
| "epoch": 0.46257748172818947, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3008940911639302e-05, |
| "loss": 9.1861, |
| "mean_token_accuracy": 0.722712657712658, |
| "mean_token_accuracy_utility": 0.5989682539682539, |
| "step": 1250 |
| }, |
| { |
| "epoch": 0.46442779165510223, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2947257720999577e-05, |
| "loss": 9.0951, |
| "mean_token_accuracy": 0.7275681725681729, |
| "mean_token_accuracy_utility": 0.5926800976800976, |
| "step": 1255 |
| }, |
| { |
| "epoch": 0.466278101582015, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2885451478993777e-05, |
| "loss": 8.995, |
| "mean_token_accuracy": 0.7236019536019535, |
| "mean_token_accuracy_utility": 0.6092063492063491, |
| "step": 1260 |
| }, |
| { |
| "epoch": 0.46812841150892776, |
| "grad_norm": 0.0, |
| "learning_rate": 1.282352476610289e-05, |
| "loss": 9.0544, |
| "mean_token_accuracy": 0.72794566648712, |
| "mean_token_accuracy_utility": 0.6005036661280262, |
| "step": 1265 |
| }, |
| { |
| "epoch": 0.4699787214358405, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2761480167837705e-05, |
| "loss": 8.9647, |
| "mean_token_accuracy": 0.7347111154505824, |
| "mean_token_accuracy_utility": 0.5976742498926504, |
| "step": 1270 |
| }, |
| { |
| "epoch": 0.4718290313627533, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2699320274630847e-05, |
| "loss": 8.9416, |
| "mean_token_accuracy": 0.7308831908831909, |
| "mean_token_accuracy_utility": 0.6019658119658118, |
| "step": 1275 |
| }, |
| { |
| "epoch": 0.47367934128966604, |
| "grad_norm": 0.0, |
| "learning_rate": 1.263704768172864e-05, |
| "loss": 9.269, |
| "mean_token_accuracy": 0.7171904649379347, |
| "mean_token_accuracy_utility": 0.5972002104426195, |
| "step": 1280 |
| }, |
| { |
| "epoch": 0.47552965121657875, |
| "grad_norm": 0.0, |
| "learning_rate": 1.257466498908276e-05, |
| "loss": 8.8439, |
| "mean_token_accuracy": 0.7327210786794961, |
| "mean_token_accuracy_utility": 0.6069544448296954, |
| "step": 1285 |
| }, |
| { |
| "epoch": 0.4773799611434915, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2512174801241657e-05, |
| "loss": 8.9788, |
| "mean_token_accuracy": 0.7342308788599895, |
| "mean_token_accuracy_utility": 0.5950491689365005, |
| "step": 1290 |
| }, |
| { |
| "epoch": 0.4792302710704043, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2449579727241834e-05, |
| "loss": 9.0183, |
| "mean_token_accuracy": 0.7289591447412184, |
| "mean_token_accuracy_utility": 0.6000520373982586, |
| "step": 1295 |
| }, |
| { |
| "epoch": 0.48108058099731704, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2386882380498918e-05, |
| "loss": 9.0128, |
| "mean_token_accuracy": 0.7272608872608874, |
| "mean_token_accuracy_utility": 0.6003663003663002, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.48108058099731704, |
| "eval_mean_token_accuracy": 0.7799949505831859, |
| "eval_reasoning_loss": 0.6924930810928345, |
| "eval_reasoning_runtime": 96.782, |
| "eval_reasoning_samples_per_second": 1.333, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.48108058099731704, |
| "eval_mean_token_accuracy": 0.5964613456552595, |
| "eval_utility_loss": 1.6131598949432373, |
| "eval_utility_runtime": 391.951, |
| "eval_utility_samples_per_second": 1.332, |
| "eval_utility_steps_per_second": 0.334, |
| "step": 1300 |
| }, |
| { |
| "epoch": 0.4829308909242298, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2324085378698529e-05, |
| "loss": 8.8228, |
| "mean_token_accuracy": 0.7398433048433048, |
| "mean_token_accuracy_utility": 0.5985531135531134, |
| "step": 1305 |
| }, |
| { |
| "epoch": 0.48478120085114257, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2261191343687e-05, |
| "loss": 9.2641, |
| "mean_token_accuracy": 0.7260418977719308, |
| "mean_token_accuracy_utility": 0.5936201988102979, |
| "step": 1310 |
| }, |
| { |
| "epoch": 0.48663151077805533, |
| "grad_norm": 0.0, |
| "learning_rate": 1.219820290136192e-05, |
| "loss": 8.9058, |
| "mean_token_accuracy": 0.7372066709566714, |
| "mean_token_accuracy_utility": 0.5975784988284987, |
| "step": 1315 |
| }, |
| { |
| "epoch": 0.4884818207049681, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2135122681562481e-05, |
| "loss": 9.0699, |
| "mean_token_accuracy": 0.7259523809523813, |
| "mean_token_accuracy_utility": 0.6002503052503051, |
| "step": 1320 |
| }, |
| { |
| "epoch": 0.49033213063188086, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2071953317959692e-05, |
| "loss": 9.1306, |
| "mean_token_accuracy": 0.7279847728410085, |
| "mean_token_accuracy_utility": 0.595906232205536, |
| "step": 1325 |
| }, |
| { |
| "epoch": 0.4921824405587936, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2008697447946421e-05, |
| "loss": 8.9368, |
| "mean_token_accuracy": 0.7298392348392354, |
| "mean_token_accuracy_utility": 0.602106227106227, |
| "step": 1330 |
| }, |
| { |
| "epoch": 0.49403275048570633, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1945357712527273e-05, |
| "loss": 9.1135, |
| "mean_token_accuracy": 0.7281198908676553, |
| "mean_token_accuracy_utility": 0.5983401365834291, |
| "step": 1335 |
| }, |
| { |
| "epoch": 0.4958830604126191, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1881936756208329e-05, |
| "loss": 9.0154, |
| "mean_token_accuracy": 0.730093632174544, |
| "mean_token_accuracy_utility": 0.5980220442647802, |
| "step": 1340 |
| }, |
| { |
| "epoch": 0.49773337033953186, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1818437226886738e-05, |
| "loss": 8.9794, |
| "mean_token_accuracy": 0.7323220560109959, |
| "mean_token_accuracy_utility": 0.600958842025661, |
| "step": 1345 |
| }, |
| { |
| "epoch": 0.4995836802664446, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1754861775740163e-05, |
| "loss": 9.1235, |
| "mean_token_accuracy": 0.7271632071632069, |
| "mean_token_accuracy_utility": 0.593137973137973, |
| "step": 1350 |
| }, |
| { |
| "epoch": 0.5014339901933574, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1691213057116082e-05, |
| "loss": 8.9625, |
| "mean_token_accuracy": 0.7331360274850324, |
| "mean_token_accuracy_utility": 0.6020820751290887, |
| "step": 1355 |
| }, |
| { |
| "epoch": 0.5032843001202701, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1627493728420978e-05, |
| "loss": 9.1329, |
| "mean_token_accuracy": 0.7279873829873832, |
| "mean_token_accuracy_utility": 0.59519536019536, |
| "step": 1360 |
| }, |
| { |
| "epoch": 0.5051346100471829, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1563706450009391e-05, |
| "loss": 9.1094, |
| "mean_token_accuracy": 0.7270329670329672, |
| "mean_token_accuracy_utility": 0.5963125763125761, |
| "step": 1365 |
| }, |
| { |
| "epoch": 0.5069849199740957, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1499853885072827e-05, |
| "loss": 8.7976, |
| "mean_token_accuracy": 0.7389690038031703, |
| "mean_token_accuracy_utility": 0.6036347281372267, |
| "step": 1370 |
| }, |
| { |
| "epoch": 0.5088352299010084, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1435938699528586e-05, |
| "loss": 9.3168, |
| "mean_token_accuracy": 0.7162845104119061, |
| "mean_token_accuracy_utility": 0.5953492577314454, |
| "step": 1375 |
| }, |
| { |
| "epoch": 0.5106855398279212, |
| "grad_norm": 0.0, |
| "learning_rate": 1.137196356190845e-05, |
| "loss": 9.1278, |
| "mean_token_accuracy": 0.7230745443179605, |
| "mean_token_accuracy_utility": 0.5952187489489966, |
| "step": 1380 |
| }, |
| { |
| "epoch": 0.5125358497548339, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1307931143247268e-05, |
| "loss": 9.1006, |
| "mean_token_accuracy": 0.7258023072208052, |
| "mean_token_accuracy_utility": 0.5966499009053943, |
| "step": 1385 |
| }, |
| { |
| "epoch": 0.5143861596817467, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1243844116971433e-05, |
| "loss": 9.0426, |
| "mean_token_accuracy": 0.7292165242165242, |
| "mean_token_accuracy_utility": 0.5985042735042734, |
| "step": 1390 |
| }, |
| { |
| "epoch": 0.5162364696086594, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1179705158787276e-05, |
| "loss": 9.2234, |
| "mean_token_accuracy": 0.7234900284900282, |
| "mean_token_accuracy_utility": 0.5950488400488398, |
| "step": 1395 |
| }, |
| { |
| "epoch": 0.5180867795355723, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1115516946569333e-05, |
| "loss": 9.0273, |
| "mean_token_accuracy": 0.7297049247049252, |
| "mean_token_accuracy_utility": 0.5977716727716725, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.5180867795355723, |
| "eval_mean_token_accuracy": 0.7801707007589364, |
| "eval_reasoning_loss": 0.6909618377685547, |
| "eval_reasoning_runtime": 96.9449, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.5180867795355723, |
| "eval_mean_token_accuracy": 0.5965632071004195, |
| "eval_utility_loss": 1.6117899417877197, |
| "eval_utility_runtime": 394.7675, |
| "eval_utility_samples_per_second": 1.322, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 1400 |
| }, |
| { |
| "epoch": 0.519937089462485, |
| "grad_norm": 0.0, |
| "learning_rate": 1.105128216024857e-05, |
| "loss": 8.9169, |
| "mean_token_accuracy": 0.733300173252294, |
| "mean_token_accuracy_utility": 0.5999615698179308, |
| "step": 1405 |
| }, |
| { |
| "epoch": 0.5217873993893978, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0987003481700456e-05, |
| "loss": 9.0822, |
| "mean_token_accuracy": 0.7299063899063901, |
| "mean_token_accuracy_utility": 0.5927838827838825, |
| "step": 1410 |
| }, |
| { |
| "epoch": 0.5236377093163105, |
| "grad_norm": 0.0, |
| "learning_rate": 1.092268359463302e-05, |
| "loss": 8.9312, |
| "mean_token_accuracy": 0.7340713195519547, |
| "mean_token_accuracy_utility": 0.5989310050503291, |
| "step": 1415 |
| }, |
| { |
| "epoch": 0.5254880192432232, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0858325184474796e-05, |
| "loss": 8.918, |
| "mean_token_accuracy": 0.7333543295660291, |
| "mean_token_accuracy_utility": 0.6027003513354486, |
| "step": 1420 |
| }, |
| { |
| "epoch": 0.527338329170136, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0793930938262689e-05, |
| "loss": 9.1173, |
| "mean_token_accuracy": 0.724044405886511, |
| "mean_token_accuracy_utility": 0.6000330955594112, |
| "step": 1425 |
| }, |
| { |
| "epoch": 0.5291886390970487, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0729503544529814e-05, |
| "loss": 8.9591, |
| "mean_token_accuracy": 0.7312947683470487, |
| "mean_token_accuracy_utility": 0.6007829619398027, |
| "step": 1430 |
| }, |
| { |
| "epoch": 0.5310389490239615, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0665045693193226e-05, |
| "loss": 8.8592, |
| "mean_token_accuracy": 0.7312561936428283, |
| "mean_token_accuracy_utility": 0.6048919020518064, |
| "step": 1435 |
| }, |
| { |
| "epoch": 0.5328892589508742, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0600560075441617e-05, |
| "loss": 8.7282, |
| "mean_token_accuracy": 0.7396167671002628, |
| "mean_token_accuracy_utility": 0.6060542085046959, |
| "step": 1440 |
| }, |
| { |
| "epoch": 0.5347395688777871, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0536049383622966e-05, |
| "loss": 9.1515, |
| "mean_token_accuracy": 0.7225046638339101, |
| "mean_token_accuracy_utility": 0.5999315739193118, |
| "step": 1445 |
| }, |
| { |
| "epoch": 0.5365898788046998, |
| "grad_norm": 0.0, |
| "learning_rate": 1.047151631113212e-05, |
| "loss": 8.9527, |
| "mean_token_accuracy": 0.7325722425722425, |
| "mean_token_accuracy_utility": 0.6006105006105005, |
| "step": 1450 |
| }, |
| { |
| "epoch": 0.5384401887316126, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0406963552298332e-05, |
| "loss": 8.8343, |
| "mean_token_accuracy": 0.7361294261294261, |
| "mean_token_accuracy_utility": 0.5969597069597068, |
| "step": 1455 |
| }, |
| { |
| "epoch": 0.5402904986585253, |
| "grad_norm": 0.0, |
| "learning_rate": 1.034239380227281e-05, |
| "loss": 8.9949, |
| "mean_token_accuracy": 0.7335113866718929, |
| "mean_token_accuracy_utility": 0.5964193859009039, |
| "step": 1460 |
| }, |
| { |
| "epoch": 0.542140808585438, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0277809756916134e-05, |
| "loss": 9.0903, |
| "mean_token_accuracy": 0.7244585025013756, |
| "mean_token_accuracy_utility": 0.6012509653795843, |
| "step": 1465 |
| }, |
| { |
| "epoch": 0.5439911185123508, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0213214112685747e-05, |
| "loss": 9.0373, |
| "mean_token_accuracy": 0.7305575905575907, |
| "mean_token_accuracy_utility": 0.599194139194139, |
| "step": 1470 |
| }, |
| { |
| "epoch": 0.5458414284392635, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0148609566523358e-05, |
| "loss": 9.204, |
| "mean_token_accuracy": 0.723257068811484, |
| "mean_token_accuracy_utility": 0.5943499610132051, |
| "step": 1475 |
| }, |
| { |
| "epoch": 0.5476917383661764, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0083998815742335e-05, |
| "loss": 9.0575, |
| "mean_token_accuracy": 0.727471872558186, |
| "mean_token_accuracy_utility": 0.6038686091275485, |
| "step": 1480 |
| }, |
| { |
| "epoch": 0.5495420482930891, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0019384557915099e-05, |
| "loss": 9.0742, |
| "mean_token_accuracy": 0.7268030118030119, |
| "mean_token_accuracy_utility": 0.5992735042735041, |
| "step": 1485 |
| }, |
| { |
| "epoch": 0.5513923582200019, |
| "grad_norm": 0.0, |
| "learning_rate": 9.9547694907605e-06, |
| "loss": 9.0977, |
| "mean_token_accuracy": 0.7272669922669925, |
| "mean_token_accuracy_utility": 0.5947924297924296, |
| "step": 1490 |
| }, |
| { |
| "epoch": 0.5532426681469146, |
| "grad_norm": 0.0, |
| "learning_rate": 9.890156312031165e-06, |
| "loss": 8.9282, |
| "mean_token_accuracy": 0.7312914134597748, |
| "mean_token_accuracy_utility": 0.5967802232853058, |
| "step": 1495 |
| }, |
| { |
| "epoch": 0.5550929780738274, |
| "grad_norm": 0.0, |
| "learning_rate": 9.825547719400889e-06, |
| "loss": 9.0219, |
| "mean_token_accuracy": 0.7328469678469678, |
| "mean_token_accuracy_utility": 0.5995909645909644, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.5550929780738274, |
| "eval_mean_token_accuracy": 0.7805592011474364, |
| "eval_reasoning_loss": 0.6900563836097717, |
| "eval_reasoning_runtime": 96.8463, |
| "eval_reasoning_samples_per_second": 1.332, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.5550929780738274, |
| "eval_mean_token_accuracy": 0.5967871165233134, |
| "eval_utility_loss": 1.6105499267578125, |
| "eval_utility_runtime": 394.3952, |
| "eval_utility_samples_per_second": 1.324, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.5569432880007401, |
| "grad_norm": 0.0, |
| "learning_rate": 9.760946410351988e-06, |
| "loss": 9.1883, |
| "mean_token_accuracy": 0.7208214819313056, |
| "mean_token_accuracy_utility": 0.597372870702341, |
| "step": 1505 |
| }, |
| { |
| "epoch": 0.5587935979276529, |
| "grad_norm": 0.0, |
| "learning_rate": 9.696355082062679e-06, |
| "loss": 9.1138, |
| "mean_token_accuracy": 0.7241961741961743, |
| "mean_token_accuracy_utility": 0.5948962148962147, |
| "step": 1510 |
| }, |
| { |
| "epoch": 0.5606439078545656, |
| "grad_norm": 0.0, |
| "learning_rate": 9.631776431294475e-06, |
| "loss": 8.9999, |
| "mean_token_accuracy": 0.7290679690679689, |
| "mean_token_accuracy_utility": 0.6008547008547007, |
| "step": 1515 |
| }, |
| { |
| "epoch": 0.5624942177814783, |
| "grad_norm": 0.0, |
| "learning_rate": 9.567213154279582e-06, |
| "loss": 8.9615, |
| "mean_token_accuracy": 0.7329792429792428, |
| "mean_token_accuracy_utility": 0.5978266178266176, |
| "step": 1520 |
| }, |
| { |
| "epoch": 0.5643445277083912, |
| "grad_norm": 0.0, |
| "learning_rate": 9.502667946608332e-06, |
| "loss": 9.0175, |
| "mean_token_accuracy": 0.7292735042735045, |
| "mean_token_accuracy_utility": 0.6007020757020756, |
| "step": 1525 |
| }, |
| { |
| "epoch": 0.5661948376353039, |
| "grad_norm": 0.0, |
| "learning_rate": 9.43814350311666e-06, |
| "loss": 9.2008, |
| "mean_token_accuracy": 0.7246682946682949, |
| "mean_token_accuracy_utility": 0.5923931623931622, |
| "step": 1530 |
| }, |
| { |
| "epoch": 0.5680451475622167, |
| "grad_norm": 0.0, |
| "learning_rate": 9.37364251777355e-06, |
| "loss": 9.1486, |
| "mean_token_accuracy": 0.7220757020757019, |
| "mean_token_accuracy_utility": 0.601196581196581, |
| "step": 1535 |
| }, |
| { |
| "epoch": 0.5698954574891294, |
| "grad_norm": 0.0, |
| "learning_rate": 9.309167683568597e-06, |
| "loss": 9.0293, |
| "mean_token_accuracy": 0.7264263428449914, |
| "mean_token_accuracy_utility": 0.6029420321979774, |
| "step": 1540 |
| }, |
| { |
| "epoch": 0.5717457674160422, |
| "grad_norm": 0.0, |
| "learning_rate": 9.244721692399545e-06, |
| "loss": 9.0166, |
| "mean_token_accuracy": 0.7314957264957265, |
| "mean_token_accuracy_utility": 0.5951221001221, |
| "step": 1545 |
| }, |
| { |
| "epoch": 0.5735960773429549, |
| "grad_norm": 0.0, |
| "learning_rate": 9.180307234959918e-06, |
| "loss": 9.0059, |
| "mean_token_accuracy": 0.7282363919968072, |
| "mean_token_accuracy_utility": 0.5989460502272949, |
| "step": 1550 |
| }, |
| { |
| "epoch": 0.5754463872698677, |
| "grad_norm": 0.0, |
| "learning_rate": 9.115927000626665e-06, |
| "loss": 9.2361, |
| "mean_token_accuracy": 0.7254497354497357, |
| "mean_token_accuracy_utility": 0.5926129426129424, |
| "step": 1555 |
| }, |
| { |
| "epoch": 0.5772966971967805, |
| "grad_norm": 0.0, |
| "learning_rate": 9.051583677347879e-06, |
| "loss": 8.9269, |
| "mean_token_accuracy": 0.7289133089133093, |
| "mean_token_accuracy_utility": 0.6035775335775334, |
| "step": 1560 |
| }, |
| { |
| "epoch": 0.5791470071236933, |
| "grad_norm": 0.0, |
| "learning_rate": 8.987279951530586e-06, |
| "loss": 9.0343, |
| "mean_token_accuracy": 0.7278734228734228, |
| "mean_token_accuracy_utility": 0.6017643467643465, |
| "step": 1565 |
| }, |
| { |
| "epoch": 0.580997317050606, |
| "grad_norm": 0.0, |
| "learning_rate": 8.923018507928564e-06, |
| "loss": 9.1951, |
| "mean_token_accuracy": 0.7223554996068621, |
| "mean_token_accuracy_utility": 0.5957966575507447, |
| "step": 1570 |
| }, |
| { |
| "epoch": 0.5828476269775187, |
| "grad_norm": 0.0, |
| "learning_rate": 8.85880202953026e-06, |
| "loss": 9.1371, |
| "mean_token_accuracy": 0.7292165242165244, |
| "mean_token_accuracy_utility": 0.5947924297924295, |
| "step": 1575 |
| }, |
| { |
| "epoch": 0.5846979369044315, |
| "grad_norm": 0.0, |
| "learning_rate": 8.79463319744677e-06, |
| "loss": 9.1351, |
| "mean_token_accuracy": 0.7204884004884006, |
| "mean_token_accuracy_utility": 0.6015873015873014, |
| "step": 1580 |
| }, |
| { |
| "epoch": 0.5865482468313442, |
| "grad_norm": 0.0, |
| "learning_rate": 8.730514690799916e-06, |
| "loss": 8.9842, |
| "mean_token_accuracy": 0.7293243793243795, |
| "mean_token_accuracy_utility": 0.5972527472527471, |
| "step": 1585 |
| }, |
| { |
| "epoch": 0.588398556758257, |
| "grad_norm": 0.0, |
| "learning_rate": 8.666449186610353e-06, |
| "loss": 8.9817, |
| "mean_token_accuracy": 0.7327126577126578, |
| "mean_token_accuracy_utility": 0.5969902319902318, |
| "step": 1590 |
| }, |
| { |
| "epoch": 0.5902488666851697, |
| "grad_norm": 0.0, |
| "learning_rate": 8.60243935968585e-06, |
| "loss": 9.0021, |
| "mean_token_accuracy": 0.730470835043707, |
| "mean_token_accuracy_utility": 0.5948923586109746, |
| "step": 1595 |
| }, |
| { |
| "epoch": 0.5920991766120826, |
| "grad_norm": 0.0, |
| "learning_rate": 8.538487882509568e-06, |
| "loss": 9.1107, |
| "mean_token_accuracy": 0.7290401450943338, |
| "mean_token_accuracy_utility": 0.5955453437079099, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.5920991766120826, |
| "eval_mean_token_accuracy": 0.7807627013509367, |
| "eval_reasoning_loss": 0.6892995834350586, |
| "eval_reasoning_runtime": 96.8995, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.5920991766120826, |
| "eval_mean_token_accuracy": 0.5968663853774469, |
| "eval_utility_loss": 1.609663486480713, |
| "eval_utility_runtime": 394.4204, |
| "eval_utility_samples_per_second": 1.323, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 1600 |
| }, |
| { |
| "epoch": 0.5939494865389953, |
| "grad_norm": 0.0, |
| "learning_rate": 8.474597425128501e-06, |
| "loss": 8.8529, |
| "mean_token_accuracy": 0.7330970559044808, |
| "mean_token_accuracy_utility": 0.6013302397525133, |
| "step": 1605 |
| }, |
| { |
| "epoch": 0.5957997964659081, |
| "grad_norm": 0.0, |
| "learning_rate": 8.410770655042003e-06, |
| "loss": 8.9083, |
| "mean_token_accuracy": 0.7297692486051941, |
| "mean_token_accuracy_utility": 0.6005409570487931, |
| "step": 1610 |
| }, |
| { |
| "epoch": 0.5976501063928208, |
| "grad_norm": 0.0, |
| "learning_rate": 8.347010237090408e-06, |
| "loss": 9.0221, |
| "mean_token_accuracy": 0.7271465604391077, |
| "mean_token_accuracy_utility": 0.6004018302794719, |
| "step": 1615 |
| }, |
| { |
| "epoch": 0.5995004163197336, |
| "grad_norm": 0.0, |
| "learning_rate": 8.283318833343773e-06, |
| "loss": 9.1715, |
| "mean_token_accuracy": 0.7213968399272225, |
| "mean_token_accuracy_utility": 0.5956923512834991, |
| "step": 1620 |
| }, |
| { |
| "epoch": 0.6013507262466463, |
| "grad_norm": 0.0, |
| "learning_rate": 8.219699102990735e-06, |
| "loss": 9.0511, |
| "mean_token_accuracy": 0.7269653867113992, |
| "mean_token_accuracy_utility": 0.6035613473794529, |
| "step": 1625 |
| }, |
| { |
| "epoch": 0.603201036173559, |
| "grad_norm": 0.0, |
| "learning_rate": 8.156153702227484e-06, |
| "loss": 9.0311, |
| "mean_token_accuracy": 0.7296581196581198, |
| "mean_token_accuracy_utility": 0.5979120879120877, |
| "step": 1630 |
| }, |
| { |
| "epoch": 0.6050513461004718, |
| "grad_norm": 0.0, |
| "learning_rate": 8.092685284146865e-06, |
| "loss": 9.1474, |
| "mean_token_accuracy": 0.7189987789987792, |
| "mean_token_accuracy_utility": 0.6026129426129424, |
| "step": 1635 |
| }, |
| { |
| "epoch": 0.6069016560273846, |
| "grad_norm": 0.0, |
| "learning_rate": 8.029296498627608e-06, |
| "loss": 9.1235, |
| "mean_token_accuracy": 0.7264573260496517, |
| "mean_token_accuracy_utility": 0.596185164962142, |
| "step": 1640 |
| }, |
| { |
| "epoch": 0.6087519659542974, |
| "grad_norm": 0.0, |
| "learning_rate": 7.965989992223693e-06, |
| "loss": 8.8565, |
| "mean_token_accuracy": 0.7333333333333332, |
| "mean_token_accuracy_utility": 0.6039804639804638, |
| "step": 1645 |
| }, |
| { |
| "epoch": 0.6106022758812101, |
| "grad_norm": 0.0, |
| "learning_rate": 7.90276840805385e-06, |
| "loss": 9.0197, |
| "mean_token_accuracy": 0.7255603035119071, |
| "mean_token_accuracy_utility": 0.6060215698763811, |
| "step": 1650 |
| }, |
| { |
| "epoch": 0.6124525858081229, |
| "grad_norm": 0.0, |
| "learning_rate": 7.839634385691214e-06, |
| "loss": 9.0088, |
| "mean_token_accuracy": 0.7287218855910447, |
| "mean_token_accuracy_utility": 0.5992059498134263, |
| "step": 1655 |
| }, |
| { |
| "epoch": 0.6143028957350356, |
| "grad_norm": 0.0, |
| "learning_rate": 7.776590561053117e-06, |
| "loss": 9.2337, |
| "mean_token_accuracy": 0.7240253293551572, |
| "mean_token_accuracy_utility": 0.5946767206662043, |
| "step": 1660 |
| }, |
| { |
| "epoch": 0.6161532056619484, |
| "grad_norm": 0.0, |
| "learning_rate": 7.713639566291028e-06, |
| "loss": 9.3142, |
| "mean_token_accuracy": 0.717545787545788, |
| "mean_token_accuracy_utility": 0.5973015873015871, |
| "step": 1665 |
| }, |
| { |
| "epoch": 0.6180035155888611, |
| "grad_norm": 0.0, |
| "learning_rate": 7.650784029680662e-06, |
| "loss": 8.7728, |
| "mean_token_accuracy": 0.7396845746845748, |
| "mean_token_accuracy_utility": 0.601166056166056, |
| "step": 1670 |
| }, |
| { |
| "epoch": 0.6198538255157738, |
| "grad_norm": 0.0, |
| "learning_rate": 7.58802657551225e-06, |
| "loss": 9.0043, |
| "mean_token_accuracy": 0.7339682539682539, |
| "mean_token_accuracy_utility": 0.5975824175824174, |
| "step": 1675 |
| }, |
| { |
| "epoch": 0.6217041354426867, |
| "grad_norm": 0.0, |
| "learning_rate": 7.52536982398097e-06, |
| "loss": 8.9088, |
| "mean_token_accuracy": 0.7292396970246757, |
| "mean_token_accuracy_utility": 0.6065225098774459, |
| "step": 1680 |
| }, |
| { |
| "epoch": 0.6235544453695994, |
| "grad_norm": 0.0, |
| "learning_rate": 7.46281639107755e-06, |
| "loss": 9.3521, |
| "mean_token_accuracy": 0.7156015329324528, |
| "mean_token_accuracy_utility": 0.5970024420024418, |
| "step": 1685 |
| }, |
| { |
| "epoch": 0.6254047552965122, |
| "grad_norm": 0.0, |
| "learning_rate": 7.400368888479048e-06, |
| "loss": 9.1863, |
| "mean_token_accuracy": 0.7197002626900602, |
| "mean_token_accuracy_utility": 0.5960604950298868, |
| "step": 1690 |
| }, |
| { |
| "epoch": 0.6272550652234249, |
| "grad_norm": 0.0, |
| "learning_rate": 7.3380299234398076e-06, |
| "loss": 9.1319, |
| "mean_token_accuracy": 0.7231155881155883, |
| "mean_token_accuracy_utility": 0.6008119658119656, |
| "step": 1695 |
| }, |
| { |
| "epoch": 0.6291053751503377, |
| "grad_norm": 0.0, |
| "learning_rate": 7.275802098682612e-06, |
| "loss": 9.248, |
| "mean_token_accuracy": 0.7194764783906085, |
| "mean_token_accuracy_utility": 0.5970863338287236, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.6291053751503377, |
| "eval_mean_token_accuracy": 0.7808200514082869, |
| "eval_reasoning_loss": 0.6887353658676147, |
| "eval_reasoning_runtime": 96.8555, |
| "eval_reasoning_samples_per_second": 1.332, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.6291053751503377, |
| "eval_mean_token_accuracy": 0.5970100187035221, |
| "eval_utility_loss": 1.6090019941329956, |
| "eval_utility_runtime": 394.3303, |
| "eval_utility_samples_per_second": 1.324, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 1700 |
| }, |
| { |
| "epoch": 0.6309556850772504, |
| "grad_norm": 0.0, |
| "learning_rate": 7.213688012290004e-06, |
| "loss": 9.0976, |
| "mean_token_accuracy": 0.7248639929781013, |
| "mean_token_accuracy_utility": 0.5982794026217275, |
| "step": 1705 |
| }, |
| { |
| "epoch": 0.6328059950041632, |
| "grad_norm": 0.0, |
| "learning_rate": 7.151690257595826e-06, |
| "loss": 9.1186, |
| "mean_token_accuracy": 0.7275518925518925, |
| "mean_token_accuracy_utility": 0.5948901098901097, |
| "step": 1710 |
| }, |
| { |
| "epoch": 0.6346563049310759, |
| "grad_norm": 0.0, |
| "learning_rate": 7.089811423076936e-06, |
| "loss": 8.8326, |
| "mean_token_accuracy": 0.7391371591371593, |
| "mean_token_accuracy_utility": 0.5992429792429791, |
| "step": 1715 |
| }, |
| { |
| "epoch": 0.6365066148579888, |
| "grad_norm": 0.0, |
| "learning_rate": 7.028054092245134e-06, |
| "loss": 9.0993, |
| "mean_token_accuracy": 0.7236191811174438, |
| "mean_token_accuracy_utility": 0.59970980220459, |
| "step": 1720 |
| }, |
| { |
| "epoch": 0.6383569247849015, |
| "grad_norm": 0.0, |
| "learning_rate": 6.966420843539321e-06, |
| "loss": 9.1275, |
| "mean_token_accuracy": 0.7229568579568578, |
| "mean_token_accuracy_utility": 0.6011172161172159, |
| "step": 1725 |
| }, |
| { |
| "epoch": 0.6402072347118142, |
| "grad_norm": 0.0, |
| "learning_rate": 6.90491425021781e-06, |
| "loss": 9.0482, |
| "mean_token_accuracy": 0.7260805860805859, |
| "mean_token_accuracy_utility": 0.6031135531135529, |
| "step": 1730 |
| }, |
| { |
| "epoch": 0.642057544638727, |
| "grad_norm": 0.0, |
| "learning_rate": 6.843536880250914e-06, |
| "loss": 8.997, |
| "mean_token_accuracy": 0.7287364473454695, |
| "mean_token_accuracy_utility": 0.5971373029643704, |
| "step": 1735 |
| }, |
| { |
| "epoch": 0.6439078545656397, |
| "grad_norm": 0.0, |
| "learning_rate": 6.7822912962137225e-06, |
| "loss": 9.2297, |
| "mean_token_accuracy": 0.7249999999999999, |
| "mean_token_accuracy_utility": 0.5973076923076921, |
| "step": 1740 |
| }, |
| { |
| "epoch": 0.6457581644925525, |
| "grad_norm": 0.0, |
| "learning_rate": 6.721180055179113e-06, |
| "loss": 9.0848, |
| "mean_token_accuracy": 0.7274682316626246, |
| "mean_token_accuracy_utility": 0.6011538461538459, |
| "step": 1745 |
| }, |
| { |
| "epoch": 0.6476084744194652, |
| "grad_norm": 0.0, |
| "learning_rate": 6.660205708610987e-06, |
| "loss": 9.0389, |
| "mean_token_accuracy": 0.7241839641839646, |
| "mean_token_accuracy_utility": 0.6047741147741146, |
| "step": 1750 |
| }, |
| { |
| "epoch": 0.649458784346378, |
| "grad_norm": 0.0, |
| "learning_rate": 6.599370802257755e-06, |
| "loss": 9.0239, |
| "mean_token_accuracy": 0.7290577940577941, |
| "mean_token_accuracy_utility": 0.5986019536019535, |
| "step": 1755 |
| }, |
| { |
| "epoch": 0.6513090942732908, |
| "grad_norm": 0.0, |
| "learning_rate": 6.5386778760460316e-06, |
| "loss": 9.1214, |
| "mean_token_accuracy": 0.7323768823768823, |
| "mean_token_accuracy_utility": 0.5923565323565322, |
| "step": 1760 |
| }, |
| { |
| "epoch": 0.6531594042002036, |
| "grad_norm": 0.0, |
| "learning_rate": 6.478129463974598e-06, |
| "loss": 9.222, |
| "mean_token_accuracy": 0.7207407407407406, |
| "mean_token_accuracy_utility": 0.5965079365079363, |
| "step": 1765 |
| }, |
| { |
| "epoch": 0.6550097141271163, |
| "grad_norm": 0.0, |
| "learning_rate": 6.417728094008613e-06, |
| "loss": 8.8173, |
| "mean_token_accuracy": 0.7358766505778551, |
| "mean_token_accuracy_utility": 0.6080328821364949, |
| "step": 1770 |
| }, |
| { |
| "epoch": 0.6568600240540291, |
| "grad_norm": 0.0, |
| "learning_rate": 6.357476287974051e-06, |
| "loss": 8.9701, |
| "mean_token_accuracy": 0.7280220653019749, |
| "mean_token_accuracy_utility": 0.600317722157451, |
| "step": 1775 |
| }, |
| { |
| "epoch": 0.6587103339809418, |
| "grad_norm": 0.0, |
| "learning_rate": 6.297376561452428e-06, |
| "loss": 8.8529, |
| "mean_token_accuracy": 0.7327289377289375, |
| "mean_token_accuracy_utility": 0.6007631257631256, |
| "step": 1780 |
| }, |
| { |
| "epoch": 0.6605606439078545, |
| "grad_norm": 0.0, |
| "learning_rate": 6.237431423675764e-06, |
| "loss": 9.0445, |
| "mean_token_accuracy": 0.7273646723646723, |
| "mean_token_accuracy_utility": 0.5977716727716726, |
| "step": 1785 |
| }, |
| { |
| "epoch": 0.6624109538347673, |
| "grad_norm": 0.0, |
| "learning_rate": 6.177643377421827e-06, |
| "loss": 8.9924, |
| "mean_token_accuracy": 0.7304242814347677, |
| "mean_token_accuracy_utility": 0.5993436623751207, |
| "step": 1790 |
| }, |
| { |
| "epoch": 0.66426126376168, |
| "grad_norm": 0.0, |
| "learning_rate": 6.118014918909633e-06, |
| "loss": 9.1133, |
| "mean_token_accuracy": 0.7264496083221498, |
| "mean_token_accuracy_utility": 0.5990191546367784, |
| "step": 1795 |
| }, |
| { |
| "epoch": 0.6661115736885929, |
| "grad_norm": 0.0, |
| "learning_rate": 6.058548537695225e-06, |
| "loss": 8.8484, |
| "mean_token_accuracy": 0.7374033374033372, |
| "mean_token_accuracy_utility": 0.6006349206349204, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.6661115736885929, |
| "eval_mean_token_accuracy": 0.781025401613637, |
| "eval_reasoning_loss": 0.6884806752204895, |
| "eval_reasoning_runtime": 96.9228, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.6661115736885929, |
| "eval_mean_token_accuracy": 0.597070000067416, |
| "eval_utility_loss": 1.608568549156189, |
| "eval_utility_runtime": 394.5456, |
| "eval_utility_samples_per_second": 1.323, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 1800 |
| }, |
| { |
| "epoch": 0.6679618836155056, |
| "grad_norm": 0.0, |
| "learning_rate": 5.999246716567737e-06, |
| "loss": 9.1499, |
| "mean_token_accuracy": 0.7232556065828146, |
| "mean_token_accuracy_utility": 0.6008535108351337, |
| "step": 1805 |
| }, |
| { |
| "epoch": 0.6698121935424184, |
| "grad_norm": 0.0, |
| "learning_rate": 5.940111931445731e-06, |
| "loss": 9.3172, |
| "mean_token_accuracy": 0.7151746107820877, |
| "mean_token_accuracy_utility": 0.5980513048737345, |
| "step": 1810 |
| }, |
| { |
| "epoch": 0.6716625034693311, |
| "grad_norm": 0.0, |
| "learning_rate": 5.881146651273825e-06, |
| "loss": 8.882, |
| "mean_token_accuracy": 0.7347517297517293, |
| "mean_token_accuracy_utility": 0.597112332112332, |
| "step": 1815 |
| }, |
| { |
| "epoch": 0.6735128133962439, |
| "grad_norm": 0.0, |
| "learning_rate": 5.822353337919616e-06, |
| "loss": 9.0456, |
| "mean_token_accuracy": 0.7298582201891534, |
| "mean_token_accuracy_utility": 0.5992694102622093, |
| "step": 1820 |
| }, |
| { |
| "epoch": 0.6753631233231566, |
| "grad_norm": 0.0, |
| "learning_rate": 5.763734446070892e-06, |
| "loss": 8.9114, |
| "mean_token_accuracy": 0.7374949124949127, |
| "mean_token_accuracy_utility": 0.5977106227106226, |
| "step": 1825 |
| }, |
| { |
| "epoch": 0.6772134332500694, |
| "grad_norm": 0.0, |
| "learning_rate": 5.705292423133133e-06, |
| "loss": 8.8681, |
| "mean_token_accuracy": 0.7373021095998522, |
| "mean_token_accuracy_utility": 0.5999283068215351, |
| "step": 1830 |
| }, |
| { |
| "epoch": 0.6790637431769821, |
| "grad_norm": 0.0, |
| "learning_rate": 5.647029709127355e-06, |
| "loss": 9.1072, |
| "mean_token_accuracy": 0.723949364647039, |
| "mean_token_accuracy_utility": 0.6000654321584553, |
| "step": 1835 |
| }, |
| { |
| "epoch": 0.6809140531038949, |
| "grad_norm": 0.0, |
| "learning_rate": 5.5889487365882065e-06, |
| "loss": 9.0623, |
| "mean_token_accuracy": 0.729563857195436, |
| "mean_token_accuracy_utility": 0.5970798499745866, |
| "step": 1840 |
| }, |
| { |
| "epoch": 0.6827643630308077, |
| "grad_norm": 0.0, |
| "learning_rate": 5.531051930462437e-06, |
| "loss": 8.8854, |
| "mean_token_accuracy": 0.7328244962937309, |
| "mean_token_accuracy_utility": 0.6045540749617786, |
| "step": 1845 |
| }, |
| { |
| "epoch": 0.6846146729577204, |
| "grad_norm": 0.0, |
| "learning_rate": 5.4733417080076325e-06, |
| "loss": 9.0463, |
| "mean_token_accuracy": 0.7316157916157915, |
| "mean_token_accuracy_utility": 0.5949084249084248, |
| "step": 1850 |
| }, |
| { |
| "epoch": 0.6864649828846332, |
| "grad_norm": 0.0, |
| "learning_rate": 5.415820478691301e-06, |
| "loss": 9.0111, |
| "mean_token_accuracy": 0.7313236757400798, |
| "mean_token_accuracy_utility": 0.5973654106146218, |
| "step": 1855 |
| }, |
| { |
| "epoch": 0.6883152928115459, |
| "grad_norm": 0.0, |
| "learning_rate": 5.358490644090263e-06, |
| "loss": 9.0408, |
| "mean_token_accuracy": 0.7249002849002852, |
| "mean_token_accuracy_utility": 0.5998168498168497, |
| "step": 1860 |
| }, |
| { |
| "epoch": 0.6901656027384587, |
| "grad_norm": 0.0, |
| "learning_rate": 5.3013545977904005e-06, |
| "loss": 9.2301, |
| "mean_token_accuracy": 0.7213519178444552, |
| "mean_token_accuracy_utility": 0.6007346302122419, |
| "step": 1865 |
| }, |
| { |
| "epoch": 0.6920159126653714, |
| "grad_norm": 0.0, |
| "learning_rate": 5.244414725286717e-06, |
| "loss": 9.1849, |
| "mean_token_accuracy": 0.7203642653642653, |
| "mean_token_accuracy_utility": 0.5976984126984125, |
| "step": 1870 |
| }, |
| { |
| "epoch": 0.6938662225922843, |
| "grad_norm": 0.0, |
| "learning_rate": 5.187673403883721e-06, |
| "loss": 8.9187, |
| "mean_token_accuracy": 0.7361884411884412, |
| "mean_token_accuracy_utility": 0.6016056166056164, |
| "step": 1875 |
| }, |
| { |
| "epoch": 0.695716532519197, |
| "grad_norm": 0.0, |
| "learning_rate": 5.131133002596199e-06, |
| "loss": 8.8313, |
| "mean_token_accuracy": 0.7363105413105413, |
| "mean_token_accuracy_utility": 0.5987362637362635, |
| "step": 1880 |
| }, |
| { |
| "epoch": 0.6975668424461097, |
| "grad_norm": 0.0, |
| "learning_rate": 5.074795882050293e-06, |
| "loss": 9.1394, |
| "mean_token_accuracy": 0.7242877492877496, |
| "mean_token_accuracy_utility": 0.5965262515262514, |
| "step": 1885 |
| }, |
| { |
| "epoch": 0.6994171523730225, |
| "grad_norm": 0.0, |
| "learning_rate": 5.018664394384942e-06, |
| "loss": 8.9517, |
| "mean_token_accuracy": 0.7339874065979038, |
| "mean_token_accuracy_utility": 0.5993492771807687, |
| "step": 1890 |
| }, |
| { |
| "epoch": 0.7012674622999352, |
| "grad_norm": 0.0, |
| "learning_rate": 4.9627408831536705e-06, |
| "loss": 8.9112, |
| "mean_token_accuracy": 0.7306364198513956, |
| "mean_token_accuracy_utility": 0.6024465000914274, |
| "step": 1895 |
| }, |
| { |
| "epoch": 0.703117772226848, |
| "grad_norm": 0.0, |
| "learning_rate": 4.907027683226761e-06, |
| "loss": 9.0244, |
| "mean_token_accuracy": 0.7323539346194272, |
| "mean_token_accuracy_utility": 0.5973908730158729, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.703117772226848, |
| "eval_mean_token_accuracy": 0.7810383516265869, |
| "eval_reasoning_loss": 0.6881875991821289, |
| "eval_reasoning_runtime": 96.8722, |
| "eval_reasoning_samples_per_second": 1.332, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.703117772226848, |
| "eval_mean_token_accuracy": 0.5970610606622797, |
| "eval_utility_loss": 1.6083298921585083, |
| "eval_utility_runtime": 394.7231, |
| "eval_utility_samples_per_second": 1.322, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 1900 |
| }, |
| { |
| "epoch": 0.7049680821537607, |
| "grad_norm": 0.0, |
| "learning_rate": 4.85152712069375e-06, |
| "loss": 8.9723, |
| "mean_token_accuracy": 0.7285877085877088, |
| "mean_token_accuracy_utility": 0.5986324786324785, |
| "step": 1905 |
| }, |
| { |
| "epoch": 0.7068183920806735, |
| "grad_norm": 0.0, |
| "learning_rate": 4.7962415127663265e-06, |
| "loss": 8.96, |
| "mean_token_accuracy": 0.731277131621959, |
| "mean_token_accuracy_utility": 0.5998020908365734, |
| "step": 1910 |
| }, |
| { |
| "epoch": 0.7086687020075862, |
| "grad_norm": 0.0, |
| "learning_rate": 4.74117316768158e-06, |
| "loss": 9.1145, |
| "mean_token_accuracy": 0.7240841449897572, |
| "mean_token_accuracy_utility": 0.5988702615870978, |
| "step": 1915 |
| }, |
| { |
| "epoch": 0.7105190119344991, |
| "grad_norm": 0.0, |
| "learning_rate": 4.686324384605629e-06, |
| "loss": 9.1024, |
| "mean_token_accuracy": 0.7259275816122108, |
| "mean_token_accuracy_utility": 0.5995165665704534, |
| "step": 1920 |
| }, |
| { |
| "epoch": 0.7123693218614118, |
| "grad_norm": 0.0, |
| "learning_rate": 4.631697453537623e-06, |
| "loss": 8.9892, |
| "mean_token_accuracy": 0.7278001628001626, |
| "mean_token_accuracy_utility": 0.6016788766788765, |
| "step": 1925 |
| }, |
| { |
| "epoch": 0.7142196317883246, |
| "grad_norm": 0.0, |
| "learning_rate": 4.577294655214144e-06, |
| "loss": 9.159, |
| "mean_token_accuracy": 0.7239780823219311, |
| "mean_token_accuracy_utility": 0.5950233800549256, |
| "step": 1930 |
| }, |
| { |
| "epoch": 0.7160699417152373, |
| "grad_norm": 0.0, |
| "learning_rate": 4.523118261013969e-06, |
| "loss": 9.1128, |
| "mean_token_accuracy": 0.7254863654863656, |
| "mean_token_accuracy_utility": 0.5989010989010988, |
| "step": 1935 |
| }, |
| { |
| "epoch": 0.71792025164215, |
| "grad_norm": 0.0, |
| "learning_rate": 4.469170532863254e-06, |
| "loss": 9.1798, |
| "mean_token_accuracy": 0.7194342694342695, |
| "mean_token_accuracy_utility": 0.6027838827838826, |
| "step": 1940 |
| }, |
| { |
| "epoch": 0.7197705615690628, |
| "grad_norm": 0.0, |
| "learning_rate": 4.415453723141081e-06, |
| "loss": 9.1113, |
| "mean_token_accuracy": 0.7226719576719579, |
| "mean_token_accuracy_utility": 0.6025213675213673, |
| "step": 1945 |
| }, |
| { |
| "epoch": 0.7216208714959755, |
| "grad_norm": 0.0, |
| "learning_rate": 4.361970074585426e-06, |
| "loss": 8.893, |
| "mean_token_accuracy": 0.7340883190883193, |
| "mean_token_accuracy_utility": 0.6007142857142855, |
| "step": 1950 |
| }, |
| { |
| "epoch": 0.7234711814228884, |
| "grad_norm": 0.0, |
| "learning_rate": 4.308721820199529e-06, |
| "loss": 9.0849, |
| "mean_token_accuracy": 0.7242613517362202, |
| "mean_token_accuracy_utility": 0.6007938232184282, |
| "step": 1955 |
| }, |
| { |
| "epoch": 0.7253214913498011, |
| "grad_norm": 0.0, |
| "learning_rate": 4.255711183158635e-06, |
| "loss": 9.0319, |
| "mean_token_accuracy": 0.7297126380190898, |
| "mean_token_accuracy_utility": 0.5993088542282089, |
| "step": 1960 |
| }, |
| { |
| "epoch": 0.7271718012767139, |
| "grad_norm": 0.0, |
| "learning_rate": 4.2029403767172175e-06, |
| "loss": 8.961, |
| "mean_token_accuracy": 0.7282519332519332, |
| "mean_token_accuracy_utility": 0.6002991452991451, |
| "step": 1965 |
| }, |
| { |
| "epoch": 0.7290221112036266, |
| "grad_norm": 0.0, |
| "learning_rate": 4.150411604116531e-06, |
| "loss": 9.0007, |
| "mean_token_accuracy": 0.7277576404641928, |
| "mean_token_accuracy_utility": 0.6035415416611976, |
| "step": 1970 |
| }, |
| { |
| "epoch": 0.7308724211305394, |
| "grad_norm": 0.0, |
| "learning_rate": 4.098127058492652e-06, |
| "loss": 9.0606, |
| "mean_token_accuracy": 0.7239991627930223, |
| "mean_token_accuracy_utility": 0.597907134288713, |
| "step": 1975 |
| }, |
| { |
| "epoch": 0.7327227310574521, |
| "grad_norm": 0.0, |
| "learning_rate": 4.0460889227849e-06, |
| "loss": 9.1335, |
| "mean_token_accuracy": 0.7252462352462354, |
| "mean_token_accuracy_utility": 0.5970207570207569, |
| "step": 1980 |
| }, |
| { |
| "epoch": 0.7345730409843649, |
| "grad_norm": 0.0, |
| "learning_rate": 3.9942993696447045e-06, |
| "loss": 8.9915, |
| "mean_token_accuracy": 0.7293443394471827, |
| "mean_token_accuracy_utility": 0.6011221514306806, |
| "step": 1985 |
| }, |
| { |
| "epoch": 0.7364233509112776, |
| "grad_norm": 0.0, |
| "learning_rate": 3.942760561344877e-06, |
| "loss": 9.0684, |
| "mean_token_accuracy": 0.7263431013431011, |
| "mean_token_accuracy_utility": 0.6023992673992672, |
| "step": 1990 |
| }, |
| { |
| "epoch": 0.7382736608381903, |
| "grad_norm": 0.0, |
| "learning_rate": 3.891474649689362e-06, |
| "loss": 9.1137, |
| "mean_token_accuracy": 0.722610907610908, |
| "mean_token_accuracy_utility": 0.5999206349206347, |
| "step": 1995 |
| }, |
| { |
| "epoch": 0.7401239707651032, |
| "grad_norm": 0.0, |
| "learning_rate": 3.840443775923365e-06, |
| "loss": 9.1423, |
| "mean_token_accuracy": 0.7254630859842294, |
| "mean_token_accuracy_utility": 0.597024178587609, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.7401239707651032, |
| "eval_mean_token_accuracy": 0.781042051630287, |
| "eval_reasoning_loss": 0.6880146861076355, |
| "eval_reasoning_runtime": 96.9419, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.7401239707651032, |
| "eval_mean_token_accuracy": 0.597131593064355, |
| "eval_utility_loss": 1.6081149578094482, |
| "eval_utility_runtime": 392.9121, |
| "eval_utility_samples_per_second": 1.329, |
| "eval_utility_steps_per_second": 0.333, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.7419742806920159, |
| "grad_norm": 0.0, |
| "learning_rate": 3.7896700706439826e-06, |
| "loss": 8.875, |
| "mean_token_accuracy": 0.7305747418031157, |
| "mean_token_accuracy_utility": 0.6066204403055612, |
| "step": 2005 |
| }, |
| { |
| "epoch": 0.7438245906189287, |
| "grad_norm": 0.0, |
| "learning_rate": 3.7391556537112282e-06, |
| "loss": 8.8609, |
| "mean_token_accuracy": 0.7333380439770871, |
| "mean_token_accuracy_utility": 0.6007467326638622, |
| "step": 2010 |
| }, |
| { |
| "epoch": 0.7456749005458414, |
| "grad_norm": 0.0, |
| "learning_rate": 3.6889026341595378e-06, |
| "loss": 8.923, |
| "mean_token_accuracy": 0.7330044532571908, |
| "mean_token_accuracy_utility": 0.5971940679522799, |
| "step": 2015 |
| }, |
| { |
| "epoch": 0.7475252104727542, |
| "grad_norm": 0.0, |
| "learning_rate": 3.6389131101096953e-06, |
| "loss": 9.0244, |
| "mean_token_accuracy": 0.7299224729869889, |
| "mean_token_accuracy_utility": 0.5997185789121271, |
| "step": 2020 |
| }, |
| { |
| "epoch": 0.7493755203996669, |
| "grad_norm": 0.0, |
| "learning_rate": 3.5891891686812597e-06, |
| "loss": 9.0936, |
| "mean_token_accuracy": 0.7233585876004996, |
| "mean_token_accuracy_utility": 0.6010683760683759, |
| "step": 2025 |
| }, |
| { |
| "epoch": 0.7512258303265797, |
| "grad_norm": 0.0, |
| "learning_rate": 3.5397328859054138e-06, |
| "loss": 9.0054, |
| "mean_token_accuracy": 0.725879120879121, |
| "mean_token_accuracy_utility": 0.6005067155067153, |
| "step": 2030 |
| }, |
| { |
| "epoch": 0.7530761402534925, |
| "grad_norm": 0.0, |
| "learning_rate": 3.490546326638273e-06, |
| "loss": 9.1937, |
| "mean_token_accuracy": 0.7211672525828895, |
| "mean_token_accuracy_utility": 0.5962197584889464, |
| "step": 2035 |
| }, |
| { |
| "epoch": 0.7549264501804053, |
| "grad_norm": 0.0, |
| "learning_rate": 3.441631544474705e-06, |
| "loss": 8.7191, |
| "mean_token_accuracy": 0.7394240944240944, |
| "mean_token_accuracy_utility": 0.6039865689865688, |
| "step": 2040 |
| }, |
| { |
| "epoch": 0.756776760107318, |
| "grad_norm": 0.0, |
| "learning_rate": 3.3929905816625653e-06, |
| "loss": 8.9666, |
| "mean_token_accuracy": 0.7285632885632887, |
| "mean_token_accuracy_utility": 0.5979609279609278, |
| "step": 2045 |
| }, |
| { |
| "epoch": 0.7586270700342307, |
| "grad_norm": 0.0, |
| "learning_rate": 3.344625469017445e-06, |
| "loss": 9.0262, |
| "mean_token_accuracy": 0.7262352462352464, |
| "mean_token_accuracy_utility": 0.6013308913308911, |
| "step": 2050 |
| }, |
| { |
| "epoch": 0.7604773799611435, |
| "grad_norm": 0.0, |
| "learning_rate": 3.2965382258378674e-06, |
| "loss": 8.975, |
| "mean_token_accuracy": 0.7364102564102567, |
| "mean_token_accuracy_utility": 0.5947985347985346, |
| "step": 2055 |
| }, |
| { |
| "epoch": 0.7623276898880562, |
| "grad_norm": 0.0, |
| "learning_rate": 3.248730859821002e-06, |
| "loss": 9.0198, |
| "mean_token_accuracy": 0.7268355718355718, |
| "mean_token_accuracy_utility": 0.6010439560439559, |
| "step": 2060 |
| }, |
| { |
| "epoch": 0.764177999814969, |
| "grad_norm": 0.0, |
| "learning_rate": 3.2012053669788136e-06, |
| "loss": 9.0429, |
| "mean_token_accuracy": 0.7261366765430781, |
| "mean_token_accuracy_utility": 0.602573643792848, |
| "step": 2065 |
| }, |
| { |
| "epoch": 0.7660283097418817, |
| "grad_norm": 0.0, |
| "learning_rate": 3.1539637315547524e-06, |
| "loss": 9.0211, |
| "mean_token_accuracy": 0.7296845746845749, |
| "mean_token_accuracy_utility": 0.5990537240537239, |
| "step": 2070 |
| }, |
| { |
| "epoch": 0.7678786196687946, |
| "grad_norm": 0.0, |
| "learning_rate": 3.1070079259408934e-06, |
| "loss": 9.0884, |
| "mean_token_accuracy": 0.7244240944240943, |
| "mean_token_accuracy_utility": 0.6026862026862025, |
| "step": 2075 |
| }, |
| { |
| "epoch": 0.7697289295957073, |
| "grad_norm": 0.0, |
| "learning_rate": 3.0603399105955966e-06, |
| "loss": 8.8626, |
| "mean_token_accuracy": 0.7286141636141638, |
| "mean_token_accuracy_utility": 0.6044993894993893, |
| "step": 2080 |
| }, |
| { |
| "epoch": 0.7715792395226201, |
| "grad_norm": 0.0, |
| "learning_rate": 3.0139616339616394e-06, |
| "loss": 9.0272, |
| "mean_token_accuracy": 0.729218559218559, |
| "mean_token_accuracy_utility": 0.5986691086691085, |
| "step": 2085 |
| }, |
| { |
| "epoch": 0.7734295494495328, |
| "grad_norm": 0.0, |
| "learning_rate": 2.9678750323848893e-06, |
| "loss": 8.9597, |
| "mean_token_accuracy": 0.7304779209099227, |
| "mean_token_accuracy_utility": 0.6020931033891085, |
| "step": 2090 |
| }, |
| { |
| "epoch": 0.7752798593764455, |
| "grad_norm": 0.0, |
| "learning_rate": 2.922082030033446e-06, |
| "loss": 9.2, |
| "mean_token_accuracy": 0.7202601377391448, |
| "mean_token_accuracy_utility": 0.5986192410562627, |
| "step": 2095 |
| }, |
| { |
| "epoch": 0.7771301693033583, |
| "grad_norm": 0.0, |
| "learning_rate": 2.8765845388172955e-06, |
| "loss": 8.9646, |
| "mean_token_accuracy": 0.733376161733021, |
| "mean_token_accuracy_utility": 0.6011297062002838, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.7771301693033583, |
| "eval_mean_token_accuracy": 0.780955101543337, |
| "eval_reasoning_loss": 0.6879242062568665, |
| "eval_reasoning_runtime": 96.9283, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.34, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.7771301693033583, |
| "eval_mean_token_accuracy": 0.597094993020471, |
| "eval_utility_loss": 1.6080310344696045, |
| "eval_utility_runtime": 394.5515, |
| "eval_utility_samples_per_second": 1.323, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 2100 |
| }, |
| { |
| "epoch": 0.778980479230271, |
| "grad_norm": 0.0, |
| "learning_rate": 2.831384458308518e-06, |
| "loss": 9.0465, |
| "mean_token_accuracy": 0.7337281237281238, |
| "mean_token_accuracy_utility": 0.5900732600732599, |
| "step": 2105 |
| }, |
| { |
| "epoch": 0.7808307891571838, |
| "grad_norm": 0.0, |
| "learning_rate": 2.7864836756619407e-06, |
| "loss": 8.9162, |
| "mean_token_accuracy": 0.7301114422120185, |
| "mean_token_accuracy_utility": 0.6023856086873371, |
| "step": 2110 |
| }, |
| { |
| "epoch": 0.7826810990840966, |
| "grad_norm": 0.0, |
| "learning_rate": 2.741884065536373e-06, |
| "loss": 8.8788, |
| "mean_token_accuracy": 0.7382376882376883, |
| "mean_token_accuracy_utility": 0.6007814407814406, |
| "step": 2115 |
| }, |
| { |
| "epoch": 0.7845314090110094, |
| "grad_norm": 0.0, |
| "learning_rate": 2.6975874900163223e-06, |
| "loss": 9.0506, |
| "mean_token_accuracy": 0.7296769310565064, |
| "mean_token_accuracy_utility": 0.6007279848667101, |
| "step": 2120 |
| }, |
| { |
| "epoch": 0.7863817189379221, |
| "grad_norm": 0.0, |
| "learning_rate": 2.6535957985342653e-06, |
| "loss": 8.9964, |
| "mean_token_accuracy": 0.7319434954199278, |
| "mean_token_accuracy_utility": 0.5980649184942146, |
| "step": 2125 |
| }, |
| { |
| "epoch": 0.7882320288648349, |
| "grad_norm": 0.0, |
| "learning_rate": 2.6099108277934105e-06, |
| "loss": 8.8214, |
| "mean_token_accuracy": 0.7358440459406955, |
| "mean_token_accuracy_utility": 0.5969948972848456, |
| "step": 2130 |
| }, |
| { |
| "epoch": 0.7900823387917476, |
| "grad_norm": 0.0, |
| "learning_rate": 2.5665344016910367e-06, |
| "loss": 8.8929, |
| "mean_token_accuracy": 0.7327838827838827, |
| "mean_token_accuracy_utility": 0.6013675213675211, |
| "step": 2135 |
| }, |
| { |
| "epoch": 0.7919326487186604, |
| "grad_norm": 0.0, |
| "learning_rate": 2.523468331242329e-06, |
| "loss": 9.0539, |
| "mean_token_accuracy": 0.7285632885632883, |
| "mean_token_accuracy_utility": 0.5989377289377288, |
| "step": 2140 |
| }, |
| { |
| "epoch": 0.7937829586455731, |
| "grad_norm": 0.0, |
| "learning_rate": 2.4807144145047734e-06, |
| "loss": 9.0461, |
| "mean_token_accuracy": 0.7222873422873425, |
| "mean_token_accuracy_utility": 0.6048840048840047, |
| "step": 2145 |
| }, |
| { |
| "epoch": 0.7956332685724858, |
| "grad_norm": 0.0, |
| "learning_rate": 2.438274436503074e-06, |
| "loss": 8.8362, |
| "mean_token_accuracy": 0.7367614924757783, |
| "mean_token_accuracy_utility": 0.5988925360353929, |
| "step": 2150 |
| }, |
| { |
| "epoch": 0.7974835784993987, |
| "grad_norm": 0.0, |
| "learning_rate": 2.396150169154644e-06, |
| "loss": 9.0655, |
| "mean_token_accuracy": 0.7241229141229141, |
| "mean_token_accuracy_utility": 0.6003540903540902, |
| "step": 2155 |
| }, |
| { |
| "epoch": 0.7993338884263114, |
| "grad_norm": 0.0, |
| "learning_rate": 2.3543433711956197e-06, |
| "loss": 8.7982, |
| "mean_token_accuracy": 0.728453629524597, |
| "mean_token_accuracy_utility": 0.6072900705029736, |
| "step": 2160 |
| }, |
| { |
| "epoch": 0.8011841983532242, |
| "grad_norm": 0.0, |
| "learning_rate": 2.3128557881074153e-06, |
| "loss": 8.9715, |
| "mean_token_accuracy": 0.7315873015873017, |
| "mean_token_accuracy_utility": 0.6013064713064711, |
| "step": 2165 |
| }, |
| { |
| "epoch": 0.8030345082801369, |
| "grad_norm": 0.0, |
| "learning_rate": 2.271689152043873e-06, |
| "loss": 9.0237, |
| "mean_token_accuracy": 0.7275396825396823, |
| "mean_token_accuracy_utility": 0.5996886446886445, |
| "step": 2170 |
| }, |
| { |
| "epoch": 0.8048848182070497, |
| "grad_norm": 0.0, |
| "learning_rate": 2.230845181758928e-06, |
| "loss": 9.165, |
| "mean_token_accuracy": 0.7184987371016784, |
| "mean_token_accuracy_utility": 0.6029076887165121, |
| "step": 2175 |
| }, |
| { |
| "epoch": 0.8067351281339624, |
| "grad_norm": 0.0, |
| "learning_rate": 2.1903255825348533e-06, |
| "loss": 8.9406, |
| "mean_token_accuracy": 0.7319841269841271, |
| "mean_token_accuracy_utility": 0.5976373626373624, |
| "step": 2180 |
| }, |
| { |
| "epoch": 0.8085854380608752, |
| "grad_norm": 0.0, |
| "learning_rate": 2.150132046111054e-06, |
| "loss": 9.0336, |
| "mean_token_accuracy": 0.7269495319495324, |
| "mean_token_accuracy_utility": 0.599200244200244, |
| "step": 2185 |
| }, |
| { |
| "epoch": 0.810435747987788, |
| "grad_norm": 0.0, |
| "learning_rate": 2.1102662506134506e-06, |
| "loss": 9.3866, |
| "mean_token_accuracy": 0.7161762311762312, |
| "mean_token_accuracy_utility": 0.594108669108669, |
| "step": 2190 |
| }, |
| { |
| "epoch": 0.8122860579147008, |
| "grad_norm": 0.0, |
| "learning_rate": 2.0707298604843964e-06, |
| "loss": 9.1594, |
| "mean_token_accuracy": 0.7277065527065525, |
| "mean_token_accuracy_utility": 0.5947191697191696, |
| "step": 2195 |
| }, |
| { |
| "epoch": 0.8141363678416135, |
| "grad_norm": 0.0, |
| "learning_rate": 2.03152452641321e-06, |
| "loss": 8.7599, |
| "mean_token_accuracy": 0.737096543892397, |
| "mean_token_accuracy_utility": 0.6030112433988019, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.8141363678416135, |
| "eval_mean_token_accuracy": 0.7808589014471371, |
| "eval_reasoning_loss": 0.6879550218582153, |
| "eval_reasoning_runtime": 96.7698, |
| "eval_reasoning_samples_per_second": 1.333, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.8141363678416135, |
| "eval_mean_token_accuracy": 0.5971002393335328, |
| "eval_utility_loss": 1.6079723834991455, |
| "eval_utility_runtime": 394.1585, |
| "eval_utility_samples_per_second": 1.324, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 2200 |
| }, |
| { |
| "epoch": 0.8159866777685262, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9926518852672294e-06, |
| "loss": 9.0107, |
| "mean_token_accuracy": 0.7287830687830689, |
| "mean_token_accuracy_utility": 0.601159951159951, |
| "step": 2205 |
| }, |
| { |
| "epoch": 0.817836987695439, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9541135600234917e-06, |
| "loss": 8.8452, |
| "mean_token_accuracy": 0.7342979242979245, |
| "mean_token_accuracy_utility": 0.6026862026862025, |
| "step": 2210 |
| }, |
| { |
| "epoch": 0.8196872976223517, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9159111597009584e-06, |
| "loss": 8.9091, |
| "mean_token_accuracy": 0.7336673575513516, |
| "mean_token_accuracy_utility": 0.5989141605661428, |
| "step": 2215 |
| }, |
| { |
| "epoch": 0.8215376075492645, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8780462792933473e-06, |
| "loss": 8.8557, |
| "mean_token_accuracy": 0.7348973134209856, |
| "mean_token_accuracy_utility": 0.6021400477110641, |
| "step": 2220 |
| }, |
| { |
| "epoch": 0.8233879174761772, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8405204997025394e-06, |
| "loss": 9.323, |
| "mean_token_accuracy": 0.7150104294632084, |
| "mean_token_accuracy_utility": 0.5961790295373653, |
| "step": 2225 |
| }, |
| { |
| "epoch": 0.82523822740309, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8033353876725578e-06, |
| "loss": 9.2692, |
| "mean_token_accuracy": 0.7178322303013259, |
| "mean_token_accuracy_utility": 0.5990400364473232, |
| "step": 2230 |
| }, |
| { |
| "epoch": 0.8270885373300028, |
| "grad_norm": 0.0, |
| "learning_rate": 1.766492495724178e-06, |
| "loss": 8.9207, |
| "mean_token_accuracy": 0.7321430686707425, |
| "mean_token_accuracy_utility": 0.6003974599804807, |
| "step": 2235 |
| }, |
| { |
| "epoch": 0.8289388472569156, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7299933620900945e-06, |
| "loss": 9.1851, |
| "mean_token_accuracy": 0.7233549223756829, |
| "mean_token_accuracy_utility": 0.5946068916691734, |
| "step": 2240 |
| }, |
| { |
| "epoch": 0.8307891571838283, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6938395106507034e-06, |
| "loss": 8.7558, |
| "mean_token_accuracy": 0.7437954490215499, |
| "mean_token_accuracy_utility": 0.601935797614101, |
| "step": 2245 |
| }, |
| { |
| "epoch": 0.832639467110741, |
| "grad_norm": 0.0, |
| "learning_rate": 1.658032450870467e-06, |
| "loss": 8.9976, |
| "mean_token_accuracy": 0.7294945986211524, |
| "mean_token_accuracy_utility": 0.5996193269989879, |
| "step": 2250 |
| }, |
| { |
| "epoch": 0.8344897770376538, |
| "grad_norm": 0.0, |
| "learning_rate": 1.622573677734911e-06, |
| "loss": 8.7895, |
| "mean_token_accuracy": 0.7367867317867318, |
| "mean_token_accuracy_utility": 0.604206349206349, |
| "step": 2255 |
| }, |
| { |
| "epoch": 0.8363400869645665, |
| "grad_norm": 0.0, |
| "learning_rate": 1.587464671688187e-06, |
| "loss": 8.9969, |
| "mean_token_accuracy": 0.7250187935690218, |
| "mean_token_accuracy_utility": 0.5988170644677492, |
| "step": 2260 |
| }, |
| { |
| "epoch": 0.8381903968914793, |
| "grad_norm": 0.0, |
| "learning_rate": 1.552706898571288e-06, |
| "loss": 9.1249, |
| "mean_token_accuracy": 0.7238214279320334, |
| "mean_token_accuracy_utility": 0.5990955414273568, |
| "step": 2265 |
| }, |
| { |
| "epoch": 0.840040706818392, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5183018095608138e-06, |
| "loss": 9.164, |
| "mean_token_accuracy": 0.7244052874018823, |
| "mean_token_accuracy_utility": 0.5974358974358973, |
| "step": 2270 |
| }, |
| { |
| "epoch": 0.8418910167453049, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4842508411084145e-06, |
| "loss": 8.8752, |
| "mean_token_accuracy": 0.7342980723849268, |
| "mean_token_accuracy_utility": 0.6048600291205923, |
| "step": 2275 |
| }, |
| { |
| "epoch": 0.8437413266722176, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4505554148807954e-06, |
| "loss": 9.1056, |
| "mean_token_accuracy": 0.7237484737484737, |
| "mean_token_accuracy_utility": 0.6017460317460316, |
| "step": 2280 |
| }, |
| { |
| "epoch": 0.8455916365991304, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4172169377003775e-06, |
| "loss": 9.031, |
| "mean_token_accuracy": 0.7274797267298371, |
| "mean_token_accuracy_utility": 0.599594247344578, |
| "step": 2285 |
| }, |
| { |
| "epoch": 0.8474419465260431, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3842368014865414e-06, |
| "loss": 9.0398, |
| "mean_token_accuracy": 0.730939441867084, |
| "mean_token_accuracy_utility": 0.5970551998381256, |
| "step": 2290 |
| }, |
| { |
| "epoch": 0.8492922564529559, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3516163831975337e-06, |
| "loss": 8.985, |
| "mean_token_accuracy": 0.7319922669922672, |
| "mean_token_accuracy_utility": 0.5975641025641024, |
| "step": 2295 |
| }, |
| { |
| "epoch": 0.8511425663798686, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3193570447729642e-06, |
| "loss": 9.2198, |
| "mean_token_accuracy": 0.7225824175824178, |
| "mean_token_accuracy_utility": 0.5988705738705737, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.8511425663798686, |
| "eval_mean_token_accuracy": 0.7810143016025372, |
| "eval_reasoning_loss": 0.6879178881645203, |
| "eval_reasoning_runtime": 96.8194, |
| "eval_reasoning_samples_per_second": 1.332, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.8511425663798686, |
| "eval_mean_token_accuracy": 0.5971697160051617, |
| "eval_utility_loss": 1.6079251766204834, |
| "eval_utility_runtime": 394.3381, |
| "eval_utility_samples_per_second": 1.324, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 2300 |
| }, |
| { |
| "epoch": 0.8529928763067813, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2874601330769488e-06, |
| "loss": 9.1937, |
| "mean_token_accuracy": 0.7225518925518926, |
| "mean_token_accuracy_utility": 0.5987057387057385, |
| "step": 2305 |
| }, |
| { |
| "epoch": 0.8548431862336942, |
| "grad_norm": 0.0, |
| "learning_rate": 1.255926979841876e-06, |
| "loss": 8.974, |
| "mean_token_accuracy": 0.7327603373240594, |
| "mean_token_accuracy_utility": 0.5962785699697356, |
| "step": 2310 |
| }, |
| { |
| "epoch": 0.8566934961606069, |
| "grad_norm": 0.0, |
| "learning_rate": 1.224758901612796e-06, |
| "loss": 8.9087, |
| "mean_token_accuracy": 0.7320350020350026, |
| "mean_token_accuracy_utility": 0.601208791208791, |
| "step": 2315 |
| }, |
| { |
| "epoch": 0.8585438060875197, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1939571996924738e-06, |
| "loss": 8.8673, |
| "mean_token_accuracy": 0.7356898132178401, |
| "mean_token_accuracy_utility": 0.6002318328159133, |
| "step": 2320 |
| }, |
| { |
| "epoch": 0.8603941160144324, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1635231600870334e-06, |
| "loss": 9.4021, |
| "mean_token_accuracy": 0.7170064426707222, |
| "mean_token_accuracy_utility": 0.5866847736776125, |
| "step": 2325 |
| }, |
| { |
| "epoch": 0.8622444259413452, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1334580534522932e-06, |
| "loss": 8.9822, |
| "mean_token_accuracy": 0.7289377289377289, |
| "mean_token_accuracy_utility": 0.6053357753357751, |
| "step": 2330 |
| }, |
| { |
| "epoch": 0.8640947358682579, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1037631350406874e-06, |
| "loss": 8.6888, |
| "mean_token_accuracy": 0.7455962555962555, |
| "mean_token_accuracy_utility": 0.6033211233211231, |
| "step": 2335 |
| }, |
| { |
| "epoch": 0.8659450457951707, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0744396446488781e-06, |
| "loss": 9.0533, |
| "mean_token_accuracy": 0.7252258852258854, |
| "mean_token_accuracy_utility": 0.5992551892551891, |
| "step": 2340 |
| }, |
| { |
| "epoch": 0.8677953557220834, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0454888065659775e-06, |
| "loss": 9.0774, |
| "mean_token_accuracy": 0.7208058608058606, |
| "mean_token_accuracy_utility": 0.6060439560439559, |
| "step": 2345 |
| }, |
| { |
| "epoch": 0.8696456656489963, |
| "grad_norm": 0.0, |
| "learning_rate": 1.0169118295224488e-06, |
| "loss": 9.245, |
| "mean_token_accuracy": 0.7202808302808299, |
| "mean_token_accuracy_utility": 0.5967521367521366, |
| "step": 2350 |
| }, |
| { |
| "epoch": 0.871495975575909, |
| "grad_norm": 0.0, |
| "learning_rate": 9.887099066396178e-07, |
| "loss": 9.1647, |
| "mean_token_accuracy": 0.7213750979860766, |
| "mean_token_accuracy_utility": 0.5978981877311231, |
| "step": 2355 |
| }, |
| { |
| "epoch": 0.8733462855028217, |
| "grad_norm": 0.0, |
| "learning_rate": 9.608842153798903e-07, |
| "loss": 9.274, |
| "mean_token_accuracy": 0.7190392628228714, |
| "mean_token_accuracy_utility": 0.5998039911548162, |
| "step": 2360 |
| }, |
| { |
| "epoch": 0.8751965954297345, |
| "grad_norm": 0.0, |
| "learning_rate": 9.33435917497556e-07, |
| "loss": 8.9613, |
| "mean_token_accuracy": 0.7310846560846561, |
| "mean_token_accuracy_utility": 0.6014590964590962, |
| "step": 2365 |
| }, |
| { |
| "epoch": 0.8770469053566472, |
| "grad_norm": 0.0, |
| "learning_rate": 9.063661589903116e-07, |
| "loss": 9.052, |
| "mean_token_accuracy": 0.7285775335775337, |
| "mean_token_accuracy_utility": 0.5965018315018313, |
| "step": 2370 |
| }, |
| { |
| "epoch": 0.87889721528356, |
| "grad_norm": 0.0, |
| "learning_rate": 8.796760700513984e-07, |
| "loss": 9.2784, |
| "mean_token_accuracy": 0.7178490028490029, |
| "mean_token_accuracy_utility": 0.5949999999999999, |
| "step": 2375 |
| }, |
| { |
| "epoch": 0.8807475252104727, |
| "grad_norm": 0.0, |
| "learning_rate": 8.533667650224253e-07, |
| "loss": 8.9513, |
| "mean_token_accuracy": 0.7304110704110705, |
| "mean_token_accuracy_utility": 0.6031379731379729, |
| "step": 2380 |
| }, |
| { |
| "epoch": 0.8825978351373855, |
| "grad_norm": 0.0, |
| "learning_rate": 8.274393423468385e-07, |
| "loss": 9.2455, |
| "mean_token_accuracy": 0.7197387780098062, |
| "mean_token_accuracy_utility": 0.6005960654091493, |
| "step": 2385 |
| }, |
| { |
| "epoch": 0.8844481450642983, |
| "grad_norm": 0.0, |
| "learning_rate": 8.018948845240538e-07, |
| "loss": 9.0428, |
| "mean_token_accuracy": 0.72497258137468, |
| "mean_token_accuracy_utility": 0.6019629211692169, |
| "step": 2390 |
| }, |
| { |
| "epoch": 0.8862984549912111, |
| "grad_norm": 0.0, |
| "learning_rate": 7.767344580642821e-07, |
| "loss": 9.0397, |
| "mean_token_accuracy": 0.7300651200651201, |
| "mean_token_accuracy_utility": 0.6017582417582416, |
| "step": 2395 |
| }, |
| { |
| "epoch": 0.8881487649181238, |
| "grad_norm": 0.0, |
| "learning_rate": 7.519591134439753e-07, |
| "loss": 9.131, |
| "mean_token_accuracy": 0.7286842075320858, |
| "mean_token_accuracy_utility": 0.5961747227183573, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.8881487649181238, |
| "eval_mean_token_accuracy": 0.7810827516709866, |
| "eval_reasoning_loss": 0.6879734992980957, |
| "eval_reasoning_runtime": 96.9041, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.8881487649181238, |
| "eval_mean_token_accuracy": 0.5971283499569707, |
| "eval_utility_loss": 1.6079424619674683, |
| "eval_utility_runtime": 391.6708, |
| "eval_utility_samples_per_second": 1.333, |
| "eval_utility_steps_per_second": 0.334, |
| "step": 2400 |
| }, |
| { |
| "epoch": 0.8899990748450366, |
| "grad_norm": 0.0, |
| "learning_rate": 7.275698850619861e-07, |
| "loss": 9.0984, |
| "mean_token_accuracy": 0.7274850947905087, |
| "mean_token_accuracy_utility": 0.5947751866914281, |
| "step": 2405 |
| }, |
| { |
| "epoch": 0.8918493847719493, |
| "grad_norm": 0.0, |
| "learning_rate": 7.035677911963712e-07, |
| "loss": 8.8454, |
| "mean_token_accuracy": 0.7364114986902991, |
| "mean_token_accuracy_utility": 0.6015666084030091, |
| "step": 2410 |
| }, |
| { |
| "epoch": 0.893699694698862, |
| "grad_norm": 0.0, |
| "learning_rate": 6.799538339618838e-07, |
| "loss": 9.1471, |
| "mean_token_accuracy": 0.7231949531949531, |
| "mean_token_accuracy_utility": 0.5955067155067153, |
| "step": 2415 |
| }, |
| { |
| "epoch": 0.8955500046257748, |
| "grad_norm": 0.0, |
| "learning_rate": 6.567289992681258e-07, |
| "loss": 8.8925, |
| "mean_token_accuracy": 0.7323782734531527, |
| "mean_token_accuracy_utility": 0.6011274943521319, |
| "step": 2420 |
| }, |
| { |
| "epoch": 0.8974003145526875, |
| "grad_norm": 0.0, |
| "learning_rate": 6.33894256778399e-07, |
| "loss": 9.0849, |
| "mean_token_accuracy": 0.7236414439911178, |
| "mean_token_accuracy_utility": 0.5968095578585793, |
| "step": 2425 |
| }, |
| { |
| "epoch": 0.8992506244796004, |
| "grad_norm": 0.0, |
| "learning_rate": 6.114505598692011e-07, |
| "loss": 9.1171, |
| "mean_token_accuracy": 0.7254361431371054, |
| "mean_token_accuracy_utility": 0.598762641865528, |
| "step": 2430 |
| }, |
| { |
| "epoch": 0.9011009344065131, |
| "grad_norm": 0.0, |
| "learning_rate": 5.893988455904387e-07, |
| "loss": 8.9488, |
| "mean_token_accuracy": 0.7336568188528976, |
| "mean_token_accuracy_utility": 0.5973807129689481, |
| "step": 2435 |
| }, |
| { |
| "epoch": 0.9029512443334259, |
| "grad_norm": 0.0, |
| "learning_rate": 5.677400346262918e-07, |
| "loss": 9.1132, |
| "mean_token_accuracy": 0.7286711436711436, |
| "mean_token_accuracy_utility": 0.5904334554334553, |
| "step": 2440 |
| }, |
| { |
| "epoch": 0.9048015542603386, |
| "grad_norm": 0.0, |
| "learning_rate": 5.464750312567835e-07, |
| "loss": 8.7514, |
| "mean_token_accuracy": 0.7423809523809525, |
| "mean_token_accuracy_utility": 0.5970451770451769, |
| "step": 2445 |
| }, |
| { |
| "epoch": 0.9066518641872514, |
| "grad_norm": 0.0, |
| "learning_rate": 5.256047233200201e-07, |
| "loss": 8.9859, |
| "mean_token_accuracy": 0.7289540089540091, |
| "mean_token_accuracy_utility": 0.5999511599511598, |
| "step": 2450 |
| }, |
| { |
| "epoch": 0.9085021741141641, |
| "grad_norm": 0.0, |
| "learning_rate": 5.051299821751254e-07, |
| "loss": 8.9483, |
| "mean_token_accuracy": 0.7281033781033782, |
| "mean_token_accuracy_utility": 0.6055555555555554, |
| "step": 2455 |
| }, |
| { |
| "epoch": 0.9103524840410768, |
| "grad_norm": 0.0, |
| "learning_rate": 4.850516626658585e-07, |
| "loss": 9.1741, |
| "mean_token_accuracy": 0.7218274318274319, |
| "mean_token_accuracy_utility": 0.6009890109890108, |
| "step": 2460 |
| }, |
| { |
| "epoch": 0.9122027939679896, |
| "grad_norm": 0.0, |
| "learning_rate": 4.653706030849214e-07, |
| "loss": 9.1882, |
| "mean_token_accuracy": 0.7259354456373112, |
| "mean_token_accuracy_utility": 0.5948148839204797, |
| "step": 2465 |
| }, |
| { |
| "epoch": 0.9140531038949024, |
| "grad_norm": 0.0, |
| "learning_rate": 4.4608762513896455e-07, |
| "loss": 9.283, |
| "mean_token_accuracy": 0.7179078370764724, |
| "mean_token_accuracy_utility": 0.5989127664186721, |
| "step": 2470 |
| }, |
| { |
| "epoch": 0.9159034138218152, |
| "grad_norm": 0.0, |
| "learning_rate": 4.2720353391427547e-07, |
| "loss": 8.9637, |
| "mean_token_accuracy": 0.7302865089168049, |
| "mean_token_accuracy_utility": 0.597294203185091, |
| "step": 2475 |
| }, |
| { |
| "epoch": 0.9177537237487279, |
| "grad_norm": 0.0, |
| "learning_rate": 4.087191178431682e-07, |
| "loss": 9.1074, |
| "mean_token_accuracy": 0.7243759381461354, |
| "mean_token_accuracy_utility": 0.5990252503358426, |
| "step": 2480 |
| }, |
| { |
| "epoch": 0.9196040336756407, |
| "grad_norm": 0.0, |
| "learning_rate": 3.9063514867105914e-07, |
| "loss": 9.3498, |
| "mean_token_accuracy": 0.7136589336589341, |
| "mean_token_accuracy_utility": 0.5957631257631256, |
| "step": 2485 |
| }, |
| { |
| "epoch": 0.9214543436025534, |
| "grad_norm": 0.0, |
| "learning_rate": 3.729523814242608e-07, |
| "loss": 8.9913, |
| "mean_token_accuracy": 0.7271752225668137, |
| "mean_token_accuracy_utility": 0.6030274992022723, |
| "step": 2490 |
| }, |
| { |
| "epoch": 0.9233046535294662, |
| "grad_norm": 0.0, |
| "learning_rate": 3.5567155437843725e-07, |
| "loss": 8.9726, |
| "mean_token_accuracy": 0.7311457061457064, |
| "mean_token_accuracy_utility": 0.5994322344322343, |
| "step": 2495 |
| }, |
| { |
| "epoch": 0.9251549634563789, |
| "grad_norm": 0.0, |
| "learning_rate": 3.3879338902779945e-07, |
| "loss": 8.7738, |
| "mean_token_accuracy": 0.739858656201325, |
| "mean_token_accuracy_utility": 0.6029337219617283, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.9251549634563789, |
| "eval_mean_token_accuracy": 0.781042051630287, |
| "eval_reasoning_loss": 0.6878708600997925, |
| "eval_reasoning_runtime": 96.8576, |
| "eval_reasoning_samples_per_second": 1.332, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.9251549634563789, |
| "eval_mean_token_accuracy": 0.5970620424863441, |
| "eval_utility_loss": 1.607924461364746, |
| "eval_utility_runtime": 392.9112, |
| "eval_utility_samples_per_second": 1.329, |
| "eval_utility_steps_per_second": 0.333, |
| "step": 2500 |
| }, |
| { |
| "epoch": 0.9270052733832918, |
| "grad_norm": 0.0, |
| "learning_rate": 3.223185900549686e-07, |
| "loss": 9.1895, |
| "mean_token_accuracy": 0.7239967439967443, |
| "mean_token_accuracy_utility": 0.5984493284493283, |
| "step": 2505 |
| }, |
| { |
| "epoch": 0.9288555833102045, |
| "grad_norm": 0.0, |
| "learning_rate": 3.0624784530156384e-07, |
| "loss": 9.0817, |
| "mean_token_accuracy": 0.7270999632909769, |
| "mean_token_accuracy_utility": 0.6017638703369111, |
| "step": 2510 |
| }, |
| { |
| "epoch": 0.9307058932371172, |
| "grad_norm": 0.0, |
| "learning_rate": 2.905818257394799e-07, |
| "loss": 8.8043, |
| "mean_token_accuracy": 0.7385490435490439, |
| "mean_token_accuracy_utility": 0.6067948717948716, |
| "step": 2515 |
| }, |
| { |
| "epoch": 0.93255620316403, |
| "grad_norm": 0.0, |
| "learning_rate": 2.753211854428728e-07, |
| "loss": 8.8651, |
| "mean_token_accuracy": 0.735611807156074, |
| "mean_token_accuracy_utility": 0.6026718073046073, |
| "step": 2520 |
| }, |
| { |
| "epoch": 0.9344065130909427, |
| "grad_norm": 0.0, |
| "learning_rate": 2.604665615608526e-07, |
| "loss": 9.0668, |
| "mean_token_accuracy": 0.7265971312315957, |
| "mean_token_accuracy_utility": 0.599126984126984, |
| "step": 2525 |
| }, |
| { |
| "epoch": 0.9362568230178555, |
| "grad_norm": 0.0, |
| "learning_rate": 2.460185742908816e-07, |
| "loss": 9.081, |
| "mean_token_accuracy": 0.7218783489215473, |
| "mean_token_accuracy_utility": 0.6042553153849101, |
| "step": 2530 |
| }, |
| { |
| "epoch": 0.9381071329447682, |
| "grad_norm": 0.0, |
| "learning_rate": 2.3197782685288385e-07, |
| "loss": 9.1348, |
| "mean_token_accuracy": 0.7230387424699087, |
| "mean_token_accuracy_utility": 0.5955875338810328, |
| "step": 2535 |
| }, |
| { |
| "epoch": 0.939957442871681, |
| "grad_norm": 0.0, |
| "learning_rate": 2.1834490546405186e-07, |
| "loss": 9.0623, |
| "mean_token_accuracy": 0.7269407830253216, |
| "mean_token_accuracy_utility": 0.5994304076840227, |
| "step": 2540 |
| }, |
| { |
| "epoch": 0.9418077527985937, |
| "grad_norm": 0.0, |
| "learning_rate": 2.0512037931437855e-07, |
| "loss": 9.0706, |
| "mean_token_accuracy": 0.7244447154718036, |
| "mean_token_accuracy_utility": 0.599915342996607, |
| "step": 2545 |
| }, |
| { |
| "epoch": 0.9436580627255066, |
| "grad_norm": 0.0, |
| "learning_rate": 1.9230480054288958e-07, |
| "loss": 8.692, |
| "mean_token_accuracy": 0.7385124135124133, |
| "mean_token_accuracy_utility": 0.6025579975579973, |
| "step": 2550 |
| }, |
| { |
| "epoch": 0.9455083726524193, |
| "grad_norm": 0.0, |
| "learning_rate": 1.7989870421459498e-07, |
| "loss": 9.0061, |
| "mean_token_accuracy": 0.732167277167277, |
| "mean_token_accuracy_utility": 0.5963064713064711, |
| "step": 2555 |
| }, |
| { |
| "epoch": 0.9473586825793321, |
| "grad_norm": 0.0, |
| "learning_rate": 1.6790260829814053e-07, |
| "loss": 9.0748, |
| "mean_token_accuracy": 0.7252570263151145, |
| "mean_token_accuracy_utility": 0.602974986149251, |
| "step": 2560 |
| }, |
| { |
| "epoch": 0.9492089925062448, |
| "grad_norm": 0.0, |
| "learning_rate": 1.5631701364419492e-07, |
| "loss": 8.6939, |
| "mean_token_accuracy": 0.7433699633699635, |
| "mean_token_accuracy_utility": 0.6029059829059827, |
| "step": 2565 |
| }, |
| { |
| "epoch": 0.9510593024331575, |
| "grad_norm": 0.0, |
| "learning_rate": 1.4514240396452438e-07, |
| "loss": 8.9223, |
| "mean_token_accuracy": 0.7321385369192769, |
| "mean_token_accuracy_utility": 0.6008478451900634, |
| "step": 2570 |
| }, |
| { |
| "epoch": 0.9529096123600703, |
| "grad_norm": 0.0, |
| "learning_rate": 1.3437924581181205e-07, |
| "loss": 8.7252, |
| "mean_token_accuracy": 0.739643289302667, |
| "mean_token_accuracy_utility": 0.6036185125966452, |
| "step": 2575 |
| }, |
| { |
| "epoch": 0.954759922286983, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2402798856016474e-07, |
| "loss": 9.071, |
| "mean_token_accuracy": 0.7212504477237702, |
| "mean_token_accuracy_utility": 0.6043007937207604, |
| "step": 2580 |
| }, |
| { |
| "epoch": 0.9566102322138959, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1408906438636236e-07, |
| "loss": 8.8308, |
| "mean_token_accuracy": 0.7345686139964822, |
| "mean_token_accuracy_utility": 0.6052198835034883, |
| "step": 2585 |
| }, |
| { |
| "epoch": 0.9584605421408086, |
| "grad_norm": 0.0, |
| "learning_rate": 1.045628882518046e-07, |
| "loss": 8.8469, |
| "mean_token_accuracy": 0.7356954766613548, |
| "mean_token_accuracy_utility": 0.5994429623405961, |
| "step": 2590 |
| }, |
| { |
| "epoch": 0.9603108520677214, |
| "grad_norm": 0.0, |
| "learning_rate": 9.544985788519589e-08, |
| "loss": 9.1513, |
| "mean_token_accuracy": 0.7218567592725047, |
| "mean_token_accuracy_utility": 0.5994774817247177, |
| "step": 2595 |
| }, |
| { |
| "epoch": 0.9621611619946341, |
| "grad_norm": 0.0, |
| "learning_rate": 8.675035376593088e-08, |
| "loss": 9.0386, |
| "mean_token_accuracy": 0.727912087912088, |
| "mean_token_accuracy_utility": 0.5978021978021977, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.9621611619946341, |
| "eval_mean_token_accuracy": 0.7811939694292637, |
| "eval_reasoning_loss": 0.6878424286842346, |
| "eval_reasoning_runtime": 96.9006, |
| "eval_reasoning_samples_per_second": 1.331, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.9621611619946341, |
| "eval_mean_token_accuracy": 0.5970533853947492, |
| "eval_utility_loss": 1.6079190969467163, |
| "eval_utility_runtime": 394.5102, |
| "eval_utility_samples_per_second": 1.323, |
| "eval_utility_steps_per_second": 0.332, |
| "step": 2600 |
| }, |
| { |
| "epoch": 0.9640114719215469, |
| "grad_norm": 0.0, |
| "learning_rate": 7.846473910821162e-08, |
| "loss": 9.0218, |
| "mean_token_accuracy": 0.7256447905017206, |
| "mean_token_accuracy_utility": 0.6029905375613278, |
| "step": 2605 |
| }, |
| { |
| "epoch": 0.9658617818484596, |
| "grad_norm": 0.0, |
| "learning_rate": 7.059335984588634e-08, |
| "loss": 8.7937, |
| "mean_token_accuracy": 0.7399796499796499, |
| "mean_token_accuracy_utility": 0.601233211233211, |
| "step": 2610 |
| }, |
| { |
| "epoch": 0.9677120917753724, |
| "grad_norm": 0.0, |
| "learning_rate": 6.313654461800322e-08, |
| "loss": 9.1209, |
| "mean_token_accuracy": 0.7189859331747004, |
| "mean_token_accuracy_utility": 0.6035267860930886, |
| "step": 2615 |
| }, |
| { |
| "epoch": 0.9695624017022851, |
| "grad_norm": 0.0, |
| "learning_rate": 5.609460475509032e-08, |
| "loss": 8.829, |
| "mean_token_accuracy": 0.7351790801790805, |
| "mean_token_accuracy_utility": 0.5996764346764345, |
| "step": 2620 |
| }, |
| { |
| "epoch": 0.9714127116291978, |
| "grad_norm": 0.0, |
| "learning_rate": 4.9467834266154756e-08, |
| "loss": 9.1389, |
| "mean_token_accuracy": 0.7233605158646347, |
| "mean_token_accuracy_utility": 0.6014490689614246, |
| "step": 2625 |
| }, |
| { |
| "epoch": 0.9732630215561107, |
| "grad_norm": 0.0, |
| "learning_rate": 4.325650982641039e-08, |
| "loss": 9.088, |
| "mean_token_accuracy": 0.7264570614570619, |
| "mean_token_accuracy_utility": 0.5965140415140413, |
| "step": 2630 |
| }, |
| { |
| "epoch": 0.9751133314830234, |
| "grad_norm": 0.0, |
| "learning_rate": 3.746089076572701e-08, |
| "loss": 9.0615, |
| "mean_token_accuracy": 0.72982091982092, |
| "mean_token_accuracy_utility": 0.5949450549450548, |
| "step": 2635 |
| }, |
| { |
| "epoch": 0.9769636414099362, |
| "grad_norm": 0.0, |
| "learning_rate": 3.208121905779904e-08, |
| "loss": 8.8327, |
| "mean_token_accuracy": 0.7363890214768694, |
| "mean_token_accuracy_utility": 0.5989717042352473, |
| "step": 2640 |
| }, |
| { |
| "epoch": 0.9788139513368489, |
| "grad_norm": 0.0, |
| "learning_rate": 2.711771931004692e-08, |
| "loss": 9.0211, |
| "mean_token_accuracy": 0.7295888210007045, |
| "mean_token_accuracy_utility": 0.5956162798519301, |
| "step": 2645 |
| }, |
| { |
| "epoch": 0.9806642612637617, |
| "grad_norm": 0.0, |
| "learning_rate": 2.257059875423795e-08, |
| "loss": 9.1585, |
| "mean_token_accuracy": 0.7192490842490845, |
| "mean_token_accuracy_utility": 0.6046459096459095, |
| "step": 2650 |
| }, |
| { |
| "epoch": 0.9825145711906744, |
| "grad_norm": 0.0, |
| "learning_rate": 1.8440047237832105e-08, |
| "loss": 8.6133, |
| "mean_token_accuracy": 0.7454552963459152, |
| "mean_token_accuracy_utility": 0.6068542895261466, |
| "step": 2655 |
| }, |
| { |
| "epoch": 0.9843648811175872, |
| "grad_norm": 0.0, |
| "learning_rate": 1.472623721606059e-08, |
| "loss": 9.0123, |
| "mean_token_accuracy": 0.7304034135997608, |
| "mean_token_accuracy_utility": 0.5988292884183293, |
| "step": 2660 |
| }, |
| { |
| "epoch": 0.9862151910445, |
| "grad_norm": 0.0, |
| "learning_rate": 1.1429323744720499e-08, |
| "loss": 9.0915, |
| "mean_token_accuracy": 0.7271082621082622, |
| "mean_token_accuracy_utility": 0.5968559218559217, |
| "step": 2665 |
| }, |
| { |
| "epoch": 0.9880655009714127, |
| "grad_norm": 0.0, |
| "learning_rate": 8.549444473702207e-09, |
| "loss": 9.204, |
| "mean_token_accuracy": 0.7196595422306323, |
| "mean_token_accuracy_utility": 0.5983803360936061, |
| "step": 2670 |
| }, |
| { |
| "epoch": 0.9899158108983255, |
| "grad_norm": 0.0, |
| "learning_rate": 6.086719641246186e-09, |
| "loss": 8.7467, |
| "mean_token_accuracy": 0.7377999749071977, |
| "mean_token_accuracy_utility": 0.6018614631831308, |
| "step": 2675 |
| }, |
| { |
| "epoch": 0.9917661208252382, |
| "grad_norm": 0.0, |
| "learning_rate": 4.041252068918145e-09, |
| "loss": 8.9601, |
| "mean_token_accuracy": 0.7360592694488095, |
| "mean_token_accuracy_utility": 0.5966515568201762, |
| "step": 2680 |
| }, |
| { |
| "epoch": 0.993616430752151, |
| "grad_norm": 0.0, |
| "learning_rate": 2.4131271573191172e-09, |
| "loss": 8.9876, |
| "mean_token_accuracy": 0.7315272844475456, |
| "mean_token_accuracy_utility": 0.5989041976649809, |
| "step": 2685 |
| }, |
| { |
| "epoch": 0.9954667406790637, |
| "grad_norm": 0.0, |
| "learning_rate": 1.2024128825172121e-09, |
| "loss": 8.8582, |
| "mean_token_accuracy": 0.7324711357336727, |
| "mean_token_accuracy_utility": 0.6053254951131054, |
| "step": 2690 |
| }, |
| { |
| "epoch": 0.9973170506059765, |
| "grad_norm": 0.0, |
| "learning_rate": 4.0915979321320967e-10, |
| "loss": 9.0239, |
| "mean_token_accuracy": 0.7261965811965816, |
| "mean_token_accuracy_utility": 0.6001404151404149, |
| "step": 2695 |
| }, |
| { |
| "epoch": 0.9991673605328892, |
| "grad_norm": 0.0, |
| "learning_rate": 3.3401008625588706e-11, |
| "loss": 9.0483, |
| "mean_token_accuracy": 0.729678686045468, |
| "mean_token_accuracy_utility": 0.5966184757188215, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.9991673605328892, |
| "eval_mean_token_accuracy": 0.7810568516450868, |
| "eval_reasoning_loss": 0.6879352927207947, |
| "eval_reasoning_runtime": 96.8646, |
| "eval_reasoning_samples_per_second": 1.332, |
| "eval_reasoning_steps_per_second": 0.341, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.9991673605328892, |
| "eval_mean_token_accuracy": 0.5971321188678533, |
| "eval_utility_loss": 1.607951045036316, |
| "eval_utility_runtime": 393.5668, |
| "eval_utility_samples_per_second": 1.326, |
| "eval_utility_steps_per_second": 0.333, |
| "step": 2700 |
| }, |
| { |
| "epoch": 0.9999074845036544, |
| "mean_token_accuracy": 0.7466931216931219, |
| "mean_token_accuracy_utility": 0.6039987789987791, |
| "step": 2702, |
| "total_flos": 0.0, |
| "train_loss": 9.352778331691649, |
| "train_runtime": 124812.4415, |
| "train_samples_per_second": 0.173, |
| "train_steps_per_second": 0.022 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 2702, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": false, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
