| { | |
| "best_global_step": 1600, | |
| "best_metric": 1.9110984802246094, | |
| "best_model_checkpoint": "./hyperclova-deobfuscation-lora/checkpoint-1600", | |
| "epoch": 2.9955555555555557, | |
| "eval_steps": 200, | |
| "global_step": 1686, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.017777777777777778, | |
| "grad_norm": 3.3687641620635986, | |
| "learning_rate": 1.8e-05, | |
| "loss": 4.1361, | |
| "mean_token_accuracy": 0.3493226237595081, | |
| "num_tokens": 22106.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.035555555555555556, | |
| "grad_norm": 2.5920090675354004, | |
| "learning_rate": 3.8e-05, | |
| "loss": 3.7165, | |
| "mean_token_accuracy": 0.4088538818061352, | |
| "num_tokens": 44943.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05333333333333334, | |
| "grad_norm": 2.5703377723693848, | |
| "learning_rate": 5.8e-05, | |
| "loss": 3.3356, | |
| "mean_token_accuracy": 0.4755532510578632, | |
| "num_tokens": 67397.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07111111111111111, | |
| "grad_norm": 1.698912262916565, | |
| "learning_rate": 7.800000000000001e-05, | |
| "loss": 2.9874, | |
| "mean_token_accuracy": 0.508383595943451, | |
| "num_tokens": 89803.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08888888888888889, | |
| "grad_norm": 1.4602556228637695, | |
| "learning_rate": 9.8e-05, | |
| "loss": 2.7854, | |
| "mean_token_accuracy": 0.5358646497130394, | |
| "num_tokens": 112364.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.10666666666666667, | |
| "grad_norm": 1.5916705131530762, | |
| "learning_rate": 0.000118, | |
| "loss": 2.6546, | |
| "mean_token_accuracy": 0.5485944993793964, | |
| "num_tokens": 134028.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12444444444444444, | |
| "grad_norm": 1.6815338134765625, | |
| "learning_rate": 0.000138, | |
| "loss": 2.606, | |
| "mean_token_accuracy": 0.5535938143730164, | |
| "num_tokens": 156703.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.14222222222222222, | |
| "grad_norm": 1.8009140491485596, | |
| "learning_rate": 0.00015800000000000002, | |
| "loss": 2.5307, | |
| "mean_token_accuracy": 0.5640750013291835, | |
| "num_tokens": 178986.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.4582855701446533, | |
| "learning_rate": 0.00017800000000000002, | |
| "loss": 2.5633, | |
| "mean_token_accuracy": 0.5567230455577373, | |
| "num_tokens": 201989.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.17777777777777778, | |
| "grad_norm": 1.663874626159668, | |
| "learning_rate": 0.00019800000000000002, | |
| "loss": 2.4672, | |
| "mean_token_accuracy": 0.5688358306884765, | |
| "num_tokens": 223936.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19555555555555557, | |
| "grad_norm": 1.6701704263687134, | |
| "learning_rate": 0.00019886506935687262, | |
| "loss": 2.4388, | |
| "mean_token_accuracy": 0.5760447531938553, | |
| "num_tokens": 246101.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.21333333333333335, | |
| "grad_norm": 1.5731302499771118, | |
| "learning_rate": 0.00019760403530895334, | |
| "loss": 2.4377, | |
| "mean_token_accuracy": 0.5711787067353725, | |
| "num_tokens": 269187.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.2311111111111111, | |
| "grad_norm": 1.4479353427886963, | |
| "learning_rate": 0.00019634300126103406, | |
| "loss": 2.3596, | |
| "mean_token_accuracy": 0.5830569051206111, | |
| "num_tokens": 291454.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.24888888888888888, | |
| "grad_norm": 1.3653457164764404, | |
| "learning_rate": 0.00019508196721311475, | |
| "loss": 2.3648, | |
| "mean_token_accuracy": 0.5807973451912403, | |
| "num_tokens": 314204.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.26666666666666666, | |
| "grad_norm": 1.4210327863693237, | |
| "learning_rate": 0.00019382093316519546, | |
| "loss": 2.3186, | |
| "mean_token_accuracy": 0.5878118917346, | |
| "num_tokens": 337167.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.28444444444444444, | |
| "grad_norm": 1.532408356666565, | |
| "learning_rate": 0.00019255989911727615, | |
| "loss": 2.3637, | |
| "mean_token_accuracy": 0.5761628717184066, | |
| "num_tokens": 360272.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3022222222222222, | |
| "grad_norm": 1.4010679721832275, | |
| "learning_rate": 0.00019129886506935687, | |
| "loss": 2.2701, | |
| "mean_token_accuracy": 0.598077318072319, | |
| "num_tokens": 382779.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.5830323696136475, | |
| "learning_rate": 0.0001900378310214376, | |
| "loss": 2.2861, | |
| "mean_token_accuracy": 0.5928302705287933, | |
| "num_tokens": 405438.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.3377777777777778, | |
| "grad_norm": 1.4623483419418335, | |
| "learning_rate": 0.00018877679697351828, | |
| "loss": 2.3192, | |
| "mean_token_accuracy": 0.5854370579123497, | |
| "num_tokens": 428660.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "grad_norm": 1.4850527048110962, | |
| "learning_rate": 0.000187515762925599, | |
| "loss": 2.256, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35555555555555557, | |
| "eval_loss": 2.254753351211548, | |
| "eval_mean_token_accuracy": 0.5965457199811935, | |
| "eval_num_tokens": 450808.0, | |
| "eval_runtime": 30.9386, | |
| "eval_samples_per_second": 32.322, | |
| "eval_steps_per_second": 8.081, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.37333333333333335, | |
| "grad_norm": 1.4195237159729004, | |
| "learning_rate": 0.00018625472887767968, | |
| "loss": 2.2607, | |
| "mean_token_accuracy": 0.594056948274374, | |
| "num_tokens": 473434.0, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.39111111111111113, | |
| "grad_norm": 1.3114796876907349, | |
| "learning_rate": 0.0001849936948297604, | |
| "loss": 2.2947, | |
| "mean_token_accuracy": 0.5898103177547455, | |
| "num_tokens": 496482.0, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.4088888888888889, | |
| "grad_norm": 1.4004285335540771, | |
| "learning_rate": 0.00018373266078184112, | |
| "loss": 2.2542, | |
| "mean_token_accuracy": 0.5970372915267944, | |
| "num_tokens": 519379.0, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.4266666666666667, | |
| "grad_norm": 1.3860116004943848, | |
| "learning_rate": 0.0001824716267339218, | |
| "loss": 2.2636, | |
| "mean_token_accuracy": 0.59425338357687, | |
| "num_tokens": 542631.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 1.3675146102905273, | |
| "learning_rate": 0.00018121059268600253, | |
| "loss": 2.2412, | |
| "mean_token_accuracy": 0.5928545072674751, | |
| "num_tokens": 565400.0, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.4622222222222222, | |
| "grad_norm": 1.4246889352798462, | |
| "learning_rate": 0.00017994955863808322, | |
| "loss": 2.1577, | |
| "mean_token_accuracy": 0.6061514511704444, | |
| "num_tokens": 588003.0, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.4046531915664673, | |
| "learning_rate": 0.00017868852459016393, | |
| "loss": 2.1862, | |
| "mean_token_accuracy": 0.6008762732148171, | |
| "num_tokens": 610974.0, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.49777777777777776, | |
| "grad_norm": 1.4038338661193848, | |
| "learning_rate": 0.00017742749054224465, | |
| "loss": 2.2219, | |
| "mean_token_accuracy": 0.5970636487007142, | |
| "num_tokens": 634093.0, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.5155555555555555, | |
| "grad_norm": 1.3291988372802734, | |
| "learning_rate": 0.00017616645649432534, | |
| "loss": 2.131, | |
| "mean_token_accuracy": 0.6172704175114632, | |
| "num_tokens": 656188.0, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 1.444318413734436, | |
| "learning_rate": 0.00017490542244640606, | |
| "loss": 2.1691, | |
| "mean_token_accuracy": 0.6066021353006363, | |
| "num_tokens": 678769.0, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5511111111111111, | |
| "grad_norm": 1.3459752798080444, | |
| "learning_rate": 0.00017364438839848675, | |
| "loss": 2.1413, | |
| "mean_token_accuracy": 0.6139265760779381, | |
| "num_tokens": 701734.0, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.5688888888888889, | |
| "grad_norm": 1.3597490787506104, | |
| "learning_rate": 0.00017238335435056746, | |
| "loss": 2.1271, | |
| "mean_token_accuracy": 0.6106095433235168, | |
| "num_tokens": 724815.0, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.5866666666666667, | |
| "grad_norm": 1.4757016897201538, | |
| "learning_rate": 0.00017112232030264818, | |
| "loss": 2.133, | |
| "mean_token_accuracy": 0.6147415205836296, | |
| "num_tokens": 746903.0, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.6044444444444445, | |
| "grad_norm": 1.4856476783752441, | |
| "learning_rate": 0.00016986128625472887, | |
| "loss": 2.1201, | |
| "mean_token_accuracy": 0.6161383926868439, | |
| "num_tokens": 768982.0, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.6222222222222222, | |
| "grad_norm": 1.2596303224563599, | |
| "learning_rate": 0.0001686002522068096, | |
| "loss": 2.1392, | |
| "mean_token_accuracy": 0.6150005847215653, | |
| "num_tokens": 791061.0, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 1.3324636220932007, | |
| "learning_rate": 0.00016733921815889028, | |
| "loss": 2.1201, | |
| "mean_token_accuracy": 0.6171063780784607, | |
| "num_tokens": 813112.0, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.6577777777777778, | |
| "grad_norm": 1.419053316116333, | |
| "learning_rate": 0.000166078184110971, | |
| "loss": 2.1237, | |
| "mean_token_accuracy": 0.6111394688487053, | |
| "num_tokens": 835469.0, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.6755555555555556, | |
| "grad_norm": 1.4507274627685547, | |
| "learning_rate": 0.0001648171500630517, | |
| "loss": 2.1387, | |
| "mean_token_accuracy": 0.604290933907032, | |
| "num_tokens": 857795.0, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.6933333333333334, | |
| "grad_norm": 1.284505844116211, | |
| "learning_rate": 0.0001635561160151324, | |
| "loss": 2.1, | |
| "mean_token_accuracy": 0.6181465938687325, | |
| "num_tokens": 879659.0, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 1.5179046392440796, | |
| "learning_rate": 0.00016229508196721312, | |
| "loss": 2.0813, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "eval_loss": 2.0953471660614014, | |
| "eval_mean_token_accuracy": 0.618859866142273, | |
| "eval_num_tokens": 902240.0, | |
| "eval_runtime": 30.5153, | |
| "eval_samples_per_second": 32.77, | |
| "eval_steps_per_second": 8.193, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7288888888888889, | |
| "grad_norm": 1.3377336263656616, | |
| "learning_rate": 0.0001610340479192938, | |
| "loss": 2.1049, | |
| "mean_token_accuracy": 0.6189975582063199, | |
| "num_tokens": 925091.0, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.7466666666666667, | |
| "grad_norm": 1.406614065170288, | |
| "learning_rate": 0.00015977301387137452, | |
| "loss": 2.1343, | |
| "mean_token_accuracy": 0.6101128354668617, | |
| "num_tokens": 948151.0, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.7644444444444445, | |
| "grad_norm": 1.3494964838027954, | |
| "learning_rate": 0.00015851197982345524, | |
| "loss": 2.0506, | |
| "mean_token_accuracy": 0.6257941454648972, | |
| "num_tokens": 970339.0, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.7822222222222223, | |
| "grad_norm": 1.3070355653762817, | |
| "learning_rate": 0.00015725094577553593, | |
| "loss": 2.0955, | |
| "mean_token_accuracy": 0.6162661850452423, | |
| "num_tokens": 993552.0, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.3954617977142334, | |
| "learning_rate": 0.00015598991172761665, | |
| "loss": 2.1119, | |
| "mean_token_accuracy": 0.6154530435800553, | |
| "num_tokens": 1015564.0, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8177777777777778, | |
| "grad_norm": 1.4015129804611206, | |
| "learning_rate": 0.00015472887767969734, | |
| "loss": 2.0153, | |
| "mean_token_accuracy": 0.6296211943030358, | |
| "num_tokens": 1037721.0, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.8355555555555556, | |
| "grad_norm": 1.41290283203125, | |
| "learning_rate": 0.00015346784363177806, | |
| "loss": 2.0914, | |
| "mean_token_accuracy": 0.6156619966030121, | |
| "num_tokens": 1060627.0, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.8533333333333334, | |
| "grad_norm": 1.3715571165084839, | |
| "learning_rate": 0.00015220680958385877, | |
| "loss": 2.0674, | |
| "mean_token_accuracy": 0.6202241629362106, | |
| "num_tokens": 1082672.0, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.8711111111111111, | |
| "grad_norm": 1.3797943592071533, | |
| "learning_rate": 0.00015094577553593946, | |
| "loss": 2.0677, | |
| "mean_token_accuracy": 0.6200241416692733, | |
| "num_tokens": 1104857.0, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 1.3080323934555054, | |
| "learning_rate": 0.00014968474148802018, | |
| "loss": 2.068, | |
| "mean_token_accuracy": 0.618759186565876, | |
| "num_tokens": 1127612.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9066666666666666, | |
| "grad_norm": 1.4698944091796875, | |
| "learning_rate": 0.0001484237074401009, | |
| "loss": 2.0736, | |
| "mean_token_accuracy": 0.6208444744348526, | |
| "num_tokens": 1150411.0, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.9244444444444444, | |
| "grad_norm": 1.3741239309310913, | |
| "learning_rate": 0.0001471626733921816, | |
| "loss": 2.0887, | |
| "mean_token_accuracy": 0.6161769673228263, | |
| "num_tokens": 1172683.0, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.9422222222222222, | |
| "grad_norm": 1.3237783908843994, | |
| "learning_rate": 0.0001459016393442623, | |
| "loss": 1.9793, | |
| "mean_token_accuracy": 0.6360917523503303, | |
| "num_tokens": 1194160.0, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.3243825435638428, | |
| "learning_rate": 0.000144640605296343, | |
| "loss": 2.0095, | |
| "mean_token_accuracy": 0.6338530048727989, | |
| "num_tokens": 1215760.0, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.9777777777777777, | |
| "grad_norm": 1.3875395059585571, | |
| "learning_rate": 0.0001433795712484237, | |
| "loss": 2.0715, | |
| "mean_token_accuracy": 0.6245882242918015, | |
| "num_tokens": 1238191.0, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.9955555555555555, | |
| "grad_norm": 1.390081524848938, | |
| "learning_rate": 0.00014211853720050443, | |
| "loss": 2.0421, | |
| "mean_token_accuracy": 0.6229756608605385, | |
| "num_tokens": 1260429.0, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.0124444444444445, | |
| "grad_norm": 1.2626862525939941, | |
| "learning_rate": 0.00014085750315258512, | |
| "loss": 1.9614, | |
| "mean_token_accuracy": 0.6359066555374547, | |
| "num_tokens": 1281232.0, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.0302222222222222, | |
| "grad_norm": 1.3941477537155151, | |
| "learning_rate": 0.00013959646910466583, | |
| "loss": 1.8782, | |
| "mean_token_accuracy": 0.6482988312840462, | |
| "num_tokens": 1304130.0, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.048, | |
| "grad_norm": 1.4020227193832397, | |
| "learning_rate": 0.00013833543505674652, | |
| "loss": 1.8602, | |
| "mean_token_accuracy": 0.65641980022192, | |
| "num_tokens": 1326753.0, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.0657777777777777, | |
| "grad_norm": 1.285709023475647, | |
| "learning_rate": 0.00013707440100882724, | |
| "loss": 1.8661, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0657777777777777, | |
| "eval_loss": 2.018383264541626, | |
| "eval_mean_token_accuracy": 0.6295498251914978, | |
| "eval_num_tokens": 1348985.0, | |
| "eval_runtime": 30.5245, | |
| "eval_samples_per_second": 32.761, | |
| "eval_steps_per_second": 8.19, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0835555555555556, | |
| "grad_norm": 1.2745097875595093, | |
| "learning_rate": 0.00013581336696090796, | |
| "loss": 1.8705, | |
| "mean_token_accuracy": 0.650456714630127, | |
| "num_tokens": 1371318.0, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.1013333333333333, | |
| "grad_norm": 1.3518744707107544, | |
| "learning_rate": 0.00013455233291298865, | |
| "loss": 1.9056, | |
| "mean_token_accuracy": 0.6455502569675445, | |
| "num_tokens": 1393816.0, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.1191111111111112, | |
| "grad_norm": 1.4413272142410278, | |
| "learning_rate": 0.00013329129886506937, | |
| "loss": 1.8994, | |
| "mean_token_accuracy": 0.6459770023822784, | |
| "num_tokens": 1416529.0, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.1368888888888888, | |
| "grad_norm": 1.3811439275741577, | |
| "learning_rate": 0.00013203026481715006, | |
| "loss": 1.9138, | |
| "mean_token_accuracy": 0.6459063500165939, | |
| "num_tokens": 1438970.0, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.1546666666666667, | |
| "grad_norm": 1.3642174005508423, | |
| "learning_rate": 0.00013076923076923077, | |
| "loss": 1.8892, | |
| "mean_token_accuracy": 0.6444340243935585, | |
| "num_tokens": 1461324.0, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.1724444444444444, | |
| "grad_norm": 1.4544634819030762, | |
| "learning_rate": 0.0001295081967213115, | |
| "loss": 1.9248, | |
| "mean_token_accuracy": 0.6391839399933815, | |
| "num_tokens": 1484246.0, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.1902222222222223, | |
| "grad_norm": 1.3715091943740845, | |
| "learning_rate": 0.00012824716267339218, | |
| "loss": 1.9105, | |
| "mean_token_accuracy": 0.6393024668097496, | |
| "num_tokens": 1507305.0, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.208, | |
| "grad_norm": 1.3897929191589355, | |
| "learning_rate": 0.0001269861286254729, | |
| "loss": 1.8714, | |
| "mean_token_accuracy": 0.6521286174654961, | |
| "num_tokens": 1529082.0, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.2257777777777779, | |
| "grad_norm": 1.3576809167861938, | |
| "learning_rate": 0.00012572509457755359, | |
| "loss": 1.8677, | |
| "mean_token_accuracy": 0.6498221024870873, | |
| "num_tokens": 1551159.0, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.2435555555555555, | |
| "grad_norm": 1.3156862258911133, | |
| "learning_rate": 0.0001244640605296343, | |
| "loss": 1.8996, | |
| "mean_token_accuracy": 0.6485181763768196, | |
| "num_tokens": 1573348.0, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.2613333333333334, | |
| "grad_norm": 1.4738845825195312, | |
| "learning_rate": 0.00012320302648171502, | |
| "loss": 1.8953, | |
| "mean_token_accuracy": 0.6465991452336312, | |
| "num_tokens": 1595989.0, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.279111111111111, | |
| "grad_norm": 1.5254158973693848, | |
| "learning_rate": 0.00012194199243379571, | |
| "loss": 1.9236, | |
| "mean_token_accuracy": 0.6474427729845047, | |
| "num_tokens": 1617895.0, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.2968888888888888, | |
| "grad_norm": 1.4867346286773682, | |
| "learning_rate": 0.00012068095838587643, | |
| "loss": 1.8766, | |
| "mean_token_accuracy": 0.6491386488080024, | |
| "num_tokens": 1640415.0, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.3146666666666667, | |
| "grad_norm": 1.3776379823684692, | |
| "learning_rate": 0.00011941992433795712, | |
| "loss": 1.8644, | |
| "mean_token_accuracy": 0.6499749347567558, | |
| "num_tokens": 1662713.0, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.3324444444444445, | |
| "grad_norm": 1.420027256011963, | |
| "learning_rate": 0.00011815889029003783, | |
| "loss": 1.8874, | |
| "mean_token_accuracy": 0.648992708325386, | |
| "num_tokens": 1684783.0, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.3502222222222222, | |
| "grad_norm": 1.356441855430603, | |
| "learning_rate": 0.00011689785624211855, | |
| "loss": 1.8937, | |
| "mean_token_accuracy": 0.6503370434045792, | |
| "num_tokens": 1706623.0, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.3679999999999999, | |
| "grad_norm": 1.4901665449142456, | |
| "learning_rate": 0.00011563682219419924, | |
| "loss": 1.9094, | |
| "mean_token_accuracy": 0.6417872324585915, | |
| "num_tokens": 1729494.0, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.3857777777777778, | |
| "grad_norm": 1.3679572343826294, | |
| "learning_rate": 0.00011437578814627996, | |
| "loss": 1.8841, | |
| "mean_token_accuracy": 0.6478032737970352, | |
| "num_tokens": 1752045.0, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.4035555555555557, | |
| "grad_norm": 1.3518086671829224, | |
| "learning_rate": 0.00011311475409836065, | |
| "loss": 1.9021, | |
| "mean_token_accuracy": 0.6460829824209213, | |
| "num_tokens": 1775601.0, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.4213333333333333, | |
| "grad_norm": 1.400870442390442, | |
| "learning_rate": 0.00011185372005044137, | |
| "loss": 1.8566, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.4213333333333333, | |
| "eval_loss": 1.9762645959854126, | |
| "eval_mean_token_accuracy": 0.6358254022598266, | |
| "eval_num_tokens": 1798633.0, | |
| "eval_runtime": 30.7115, | |
| "eval_samples_per_second": 32.561, | |
| "eval_steps_per_second": 8.14, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.439111111111111, | |
| "grad_norm": 1.4487619400024414, | |
| "learning_rate": 0.00011059268600252208, | |
| "loss": 1.8417, | |
| "mean_token_accuracy": 0.6496566243469715, | |
| "num_tokens": 1820656.0, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.456888888888889, | |
| "grad_norm": 1.4507944583892822, | |
| "learning_rate": 0.00010933165195460277, | |
| "loss": 1.8829, | |
| "mean_token_accuracy": 0.647446171939373, | |
| "num_tokens": 1842871.0, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.4746666666666668, | |
| "grad_norm": 1.3563170433044434, | |
| "learning_rate": 0.00010807061790668349, | |
| "loss": 1.8508, | |
| "mean_token_accuracy": 0.6544376760721207, | |
| "num_tokens": 1865652.0, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.4924444444444445, | |
| "grad_norm": 1.366861343383789, | |
| "learning_rate": 0.00010680958385876418, | |
| "loss": 1.8756, | |
| "mean_token_accuracy": 0.648781743645668, | |
| "num_tokens": 1888455.0, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.5102222222222221, | |
| "grad_norm": 1.5031019449234009, | |
| "learning_rate": 0.0001055485498108449, | |
| "loss": 1.8461, | |
| "mean_token_accuracy": 0.6583632439374923, | |
| "num_tokens": 1910676.0, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.528, | |
| "grad_norm": 1.5248113870620728, | |
| "learning_rate": 0.00010428751576292561, | |
| "loss": 1.8857, | |
| "mean_token_accuracy": 0.6470584884285927, | |
| "num_tokens": 1933253.0, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.545777777777778, | |
| "grad_norm": 1.4354236125946045, | |
| "learning_rate": 0.0001030264817150063, | |
| "loss": 1.892, | |
| "mean_token_accuracy": 0.6445729210972786, | |
| "num_tokens": 1955820.0, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.5635555555555556, | |
| "grad_norm": 1.4288746118545532, | |
| "learning_rate": 0.00010176544766708702, | |
| "loss": 1.878, | |
| "mean_token_accuracy": 0.6476826578378677, | |
| "num_tokens": 1978120.0, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.5813333333333333, | |
| "grad_norm": 1.433902382850647, | |
| "learning_rate": 0.00010050441361916771, | |
| "loss": 1.8199, | |
| "mean_token_accuracy": 0.6561270505189896, | |
| "num_tokens": 2000508.0, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.5991111111111111, | |
| "grad_norm": 1.332987904548645, | |
| "learning_rate": 9.924337957124843e-05, | |
| "loss": 1.8555, | |
| "mean_token_accuracy": 0.6499203637242317, | |
| "num_tokens": 2023356.0, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.616888888888889, | |
| "grad_norm": 1.3830794095993042, | |
| "learning_rate": 9.798234552332913e-05, | |
| "loss": 1.8108, | |
| "mean_token_accuracy": 0.6618377715349197, | |
| "num_tokens": 2045965.0, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.6346666666666667, | |
| "grad_norm": 1.3988080024719238, | |
| "learning_rate": 9.672131147540983e-05, | |
| "loss": 1.8791, | |
| "mean_token_accuracy": 0.6456288158893585, | |
| "num_tokens": 2069108.0, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.6524444444444444, | |
| "grad_norm": 1.398549199104309, | |
| "learning_rate": 9.546027742749055e-05, | |
| "loss": 1.8885, | |
| "mean_token_accuracy": 0.6464410901069642, | |
| "num_tokens": 2091755.0, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.6702222222222223, | |
| "grad_norm": 1.5381189584732056, | |
| "learning_rate": 9.419924337957125e-05, | |
| "loss": 1.853, | |
| "mean_token_accuracy": 0.6534279838204384, | |
| "num_tokens": 2114496.0, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.688, | |
| "grad_norm": 1.4101791381835938, | |
| "learning_rate": 9.293820933165196e-05, | |
| "loss": 1.8696, | |
| "mean_token_accuracy": 0.6475102975964546, | |
| "num_tokens": 2137041.0, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.7057777777777776, | |
| "grad_norm": 1.496955156326294, | |
| "learning_rate": 9.167717528373266e-05, | |
| "loss": 1.8752, | |
| "mean_token_accuracy": 0.6469297721982002, | |
| "num_tokens": 2159711.0, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.7235555555555555, | |
| "grad_norm": 1.4269644021987915, | |
| "learning_rate": 9.041614123581336e-05, | |
| "loss": 1.8773, | |
| "mean_token_accuracy": 0.6508476585149765, | |
| "num_tokens": 2181675.0, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.7413333333333334, | |
| "grad_norm": 1.4438400268554688, | |
| "learning_rate": 8.915510718789408e-05, | |
| "loss": 1.8433, | |
| "mean_token_accuracy": 0.656335887312889, | |
| "num_tokens": 2204671.0, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.759111111111111, | |
| "grad_norm": 1.3846147060394287, | |
| "learning_rate": 8.789407313997479e-05, | |
| "loss": 1.8649, | |
| "mean_token_accuracy": 0.6451441869139671, | |
| "num_tokens": 2227866.0, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.7768888888888887, | |
| "grad_norm": 1.5432794094085693, | |
| "learning_rate": 8.663303909205549e-05, | |
| "loss": 1.8435, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.7768888888888887, | |
| "eval_loss": 1.9385051727294922, | |
| "eval_mean_token_accuracy": 0.6399562013149261, | |
| "eval_num_tokens": 2250379.0, | |
| "eval_runtime": 30.7428, | |
| "eval_samples_per_second": 32.528, | |
| "eval_steps_per_second": 8.132, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.7946666666666666, | |
| "grad_norm": 1.4225345849990845, | |
| "learning_rate": 8.537200504413619e-05, | |
| "loss": 1.8767, | |
| "mean_token_accuracy": 0.6512193940579891, | |
| "num_tokens": 2272999.0, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.8124444444444445, | |
| "grad_norm": 1.3732675313949585, | |
| "learning_rate": 8.41109709962169e-05, | |
| "loss": 1.845, | |
| "mean_token_accuracy": 0.6550421059131623, | |
| "num_tokens": 2295110.0, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.8302222222222222, | |
| "grad_norm": 1.3867266178131104, | |
| "learning_rate": 8.284993694829761e-05, | |
| "loss": 1.8236, | |
| "mean_token_accuracy": 0.6565809994935989, | |
| "num_tokens": 2317432.0, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.8479999999999999, | |
| "grad_norm": 1.3360997438430786, | |
| "learning_rate": 8.158890290037832e-05, | |
| "loss": 1.8642, | |
| "mean_token_accuracy": 0.6463637053966522, | |
| "num_tokens": 2340305.0, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.8657777777777778, | |
| "grad_norm": 1.4467201232910156, | |
| "learning_rate": 8.032786885245902e-05, | |
| "loss": 1.8661, | |
| "mean_token_accuracy": 0.653101560473442, | |
| "num_tokens": 2362746.0, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.8835555555555556, | |
| "grad_norm": 1.3943202495574951, | |
| "learning_rate": 7.906683480453972e-05, | |
| "loss": 1.8399, | |
| "mean_token_accuracy": 0.6510771587491035, | |
| "num_tokens": 2385647.0, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.9013333333333333, | |
| "grad_norm": 1.4589335918426514, | |
| "learning_rate": 7.780580075662043e-05, | |
| "loss": 1.8522, | |
| "mean_token_accuracy": 0.6478627189993859, | |
| "num_tokens": 2408732.0, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.919111111111111, | |
| "grad_norm": 1.5676307678222656, | |
| "learning_rate": 7.654476670870114e-05, | |
| "loss": 1.8316, | |
| "mean_token_accuracy": 0.656819324195385, | |
| "num_tokens": 2431137.0, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.9368888888888889, | |
| "grad_norm": 1.3882263898849487, | |
| "learning_rate": 7.528373266078185e-05, | |
| "loss": 1.7965, | |
| "mean_token_accuracy": 0.6616187065839767, | |
| "num_tokens": 2453517.0, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.9546666666666668, | |
| "grad_norm": 1.5195387601852417, | |
| "learning_rate": 7.402269861286255e-05, | |
| "loss": 1.8346, | |
| "mean_token_accuracy": 0.6537548035383225, | |
| "num_tokens": 2475737.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.9724444444444444, | |
| "grad_norm": 1.3485065698623657, | |
| "learning_rate": 7.276166456494325e-05, | |
| "loss": 1.835, | |
| "mean_token_accuracy": 0.6512499779462815, | |
| "num_tokens": 2497880.0, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.9902222222222221, | |
| "grad_norm": 1.5932726860046387, | |
| "learning_rate": 7.150063051702396e-05, | |
| "loss": 1.8432, | |
| "mean_token_accuracy": 0.6523947417736053, | |
| "num_tokens": 2519974.0, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 2.007111111111111, | |
| "grad_norm": 1.3682020902633667, | |
| "learning_rate": 7.023959646910467e-05, | |
| "loss": 1.7383, | |
| "mean_token_accuracy": 0.6761166123967421, | |
| "num_tokens": 2540296.0, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.024888888888889, | |
| "grad_norm": 1.4417686462402344, | |
| "learning_rate": 6.897856242118538e-05, | |
| "loss": 1.697, | |
| "mean_token_accuracy": 0.6766574695706368, | |
| "num_tokens": 2561919.0, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.042666666666667, | |
| "grad_norm": 1.375542163848877, | |
| "learning_rate": 6.771752837326608e-05, | |
| "loss": 1.7361, | |
| "mean_token_accuracy": 0.6702861517667771, | |
| "num_tokens": 2584915.0, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.0604444444444443, | |
| "grad_norm": 1.4783133268356323, | |
| "learning_rate": 6.645649432534678e-05, | |
| "loss": 1.6927, | |
| "mean_token_accuracy": 0.6790769457817077, | |
| "num_tokens": 2606857.0, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.078222222222222, | |
| "grad_norm": 1.5346624851226807, | |
| "learning_rate": 6.519546027742749e-05, | |
| "loss": 1.6938, | |
| "mean_token_accuracy": 0.6730351656675339, | |
| "num_tokens": 2629737.0, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.096, | |
| "grad_norm": 1.430298089981079, | |
| "learning_rate": 6.39344262295082e-05, | |
| "loss": 1.6476, | |
| "mean_token_accuracy": 0.6835471093654633, | |
| "num_tokens": 2651936.0, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.113777777777778, | |
| "grad_norm": 1.4968252182006836, | |
| "learning_rate": 6.267339218158891e-05, | |
| "loss": 1.7242, | |
| "mean_token_accuracy": 0.6697919353842735, | |
| "num_tokens": 2675241.0, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.1315555555555554, | |
| "grad_norm": 1.3892192840576172, | |
| "learning_rate": 6.141235813366961e-05, | |
| "loss": 1.6916, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.1315555555555554, | |
| "eval_loss": 1.934017539024353, | |
| "eval_mean_token_accuracy": 0.6421641361713409, | |
| "eval_num_tokens": 2698219.0, | |
| "eval_runtime": 30.3979, | |
| "eval_samples_per_second": 32.897, | |
| "eval_steps_per_second": 8.224, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.1493333333333333, | |
| "grad_norm": 1.4893920421600342, | |
| "learning_rate": 6.0151324085750316e-05, | |
| "loss": 1.7047, | |
| "mean_token_accuracy": 0.6755686655640603, | |
| "num_tokens": 2721580.0, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 2.167111111111111, | |
| "grad_norm": 1.50564444065094, | |
| "learning_rate": 5.889029003783102e-05, | |
| "loss": 1.7058, | |
| "mean_token_accuracy": 0.6746152400970459, | |
| "num_tokens": 2744110.0, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 2.1848888888888887, | |
| "grad_norm": 1.461367130279541, | |
| "learning_rate": 5.7629255989911736e-05, | |
| "loss": 1.684, | |
| "mean_token_accuracy": 0.6813082948327065, | |
| "num_tokens": 2765844.0, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 2.2026666666666666, | |
| "grad_norm": 1.553553819656372, | |
| "learning_rate": 5.636822194199244e-05, | |
| "loss": 1.6848, | |
| "mean_token_accuracy": 0.677204079926014, | |
| "num_tokens": 2788070.0, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 2.2204444444444444, | |
| "grad_norm": 1.4453001022338867, | |
| "learning_rate": 5.510718789407314e-05, | |
| "loss": 1.7182, | |
| "mean_token_accuracy": 0.6765570789575577, | |
| "num_tokens": 2810964.0, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.2382222222222223, | |
| "grad_norm": 1.5605733394622803, | |
| "learning_rate": 5.384615384615385e-05, | |
| "loss": 1.6772, | |
| "mean_token_accuracy": 0.678339496254921, | |
| "num_tokens": 2833176.0, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 2.2560000000000002, | |
| "grad_norm": 1.514710783958435, | |
| "learning_rate": 5.258511979823455e-05, | |
| "loss": 1.7192, | |
| "mean_token_accuracy": 0.6710417225956917, | |
| "num_tokens": 2855440.0, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 2.2737777777777777, | |
| "grad_norm": 1.510834813117981, | |
| "learning_rate": 5.132408575031527e-05, | |
| "loss": 1.6599, | |
| "mean_token_accuracy": 0.6822926640510559, | |
| "num_tokens": 2877713.0, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 2.2915555555555556, | |
| "grad_norm": 1.3550519943237305, | |
| "learning_rate": 5.006305170239597e-05, | |
| "loss": 1.7072, | |
| "mean_token_accuracy": 0.6754195600748062, | |
| "num_tokens": 2899934.0, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 2.3093333333333335, | |
| "grad_norm": 1.5602107048034668, | |
| "learning_rate": 4.8802017654476674e-05, | |
| "loss": 1.7111, | |
| "mean_token_accuracy": 0.6702851369976998, | |
| "num_tokens": 2923244.0, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.327111111111111, | |
| "grad_norm": 1.5889501571655273, | |
| "learning_rate": 4.754098360655738e-05, | |
| "loss": 1.6858, | |
| "mean_token_accuracy": 0.6781487062573432, | |
| "num_tokens": 2945580.0, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 2.344888888888889, | |
| "grad_norm": 1.4793872833251953, | |
| "learning_rate": 4.627994955863809e-05, | |
| "loss": 1.6799, | |
| "mean_token_accuracy": 0.6740961462259293, | |
| "num_tokens": 2969470.0, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 2.3626666666666667, | |
| "grad_norm": 1.6188234090805054, | |
| "learning_rate": 4.501891551071879e-05, | |
| "loss": 1.6838, | |
| "mean_token_accuracy": 0.6732241719961166, | |
| "num_tokens": 2991982.0, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 2.3804444444444446, | |
| "grad_norm": 1.474108338356018, | |
| "learning_rate": 4.37578814627995e-05, | |
| "loss": 1.7024, | |
| "mean_token_accuracy": 0.675683145225048, | |
| "num_tokens": 3014206.0, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 2.398222222222222, | |
| "grad_norm": 1.4645053148269653, | |
| "learning_rate": 4.2496847414880205e-05, | |
| "loss": 1.6564, | |
| "mean_token_accuracy": 0.6787498995661736, | |
| "num_tokens": 3036651.0, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.416, | |
| "grad_norm": 1.498451828956604, | |
| "learning_rate": 4.1235813366960915e-05, | |
| "loss": 1.694, | |
| "mean_token_accuracy": 0.6752896070480346, | |
| "num_tokens": 3058745.0, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 2.433777777777778, | |
| "grad_norm": 1.5558826923370361, | |
| "learning_rate": 3.997477931904162e-05, | |
| "loss": 1.7106, | |
| "mean_token_accuracy": 0.6742441862821579, | |
| "num_tokens": 3081726.0, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.4515555555555557, | |
| "grad_norm": 1.5872586965560913, | |
| "learning_rate": 3.871374527112232e-05, | |
| "loss": 1.6848, | |
| "mean_token_accuracy": 0.6764188826084137, | |
| "num_tokens": 3104292.0, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 2.469333333333333, | |
| "grad_norm": 1.551299810409546, | |
| "learning_rate": 3.745271122320303e-05, | |
| "loss": 1.6909, | |
| "mean_token_accuracy": 0.6762390181422233, | |
| "num_tokens": 3126334.0, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 2.487111111111111, | |
| "grad_norm": 1.57632315158844, | |
| "learning_rate": 3.6191677175283736e-05, | |
| "loss": 1.7113, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.487111111111111, | |
| "eval_loss": 1.922593355178833, | |
| "eval_mean_token_accuracy": 0.6445384075641633, | |
| "eval_num_tokens": 3149845.0, | |
| "eval_runtime": 30.0209, | |
| "eval_samples_per_second": 33.31, | |
| "eval_steps_per_second": 8.328, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.504888888888889, | |
| "grad_norm": 1.487930178642273, | |
| "learning_rate": 3.4930643127364446e-05, | |
| "loss": 1.7087, | |
| "mean_token_accuracy": 0.6737867616117, | |
| "num_tokens": 3172117.0, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 2.522666666666667, | |
| "grad_norm": 1.5210868120193481, | |
| "learning_rate": 3.366960907944515e-05, | |
| "loss": 1.7009, | |
| "mean_token_accuracy": 0.6799842938780785, | |
| "num_tokens": 3194261.0, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 2.5404444444444443, | |
| "grad_norm": 1.6295726299285889, | |
| "learning_rate": 3.240857503152585e-05, | |
| "loss": 1.6027, | |
| "mean_token_accuracy": 0.6899775773286819, | |
| "num_tokens": 3216455.0, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 2.558222222222222, | |
| "grad_norm": 1.561673879623413, | |
| "learning_rate": 3.114754098360656e-05, | |
| "loss": 1.7273, | |
| "mean_token_accuracy": 0.6699303150177002, | |
| "num_tokens": 3238359.0, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 2.576, | |
| "grad_norm": 1.5006392002105713, | |
| "learning_rate": 2.9886506935687263e-05, | |
| "loss": 1.7243, | |
| "mean_token_accuracy": 0.6692202746868133, | |
| "num_tokens": 3261568.0, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.5937777777777775, | |
| "grad_norm": 1.602378249168396, | |
| "learning_rate": 2.8625472887767974e-05, | |
| "loss": 1.7255, | |
| "mean_token_accuracy": 0.6683675542473793, | |
| "num_tokens": 3284517.0, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 2.6115555555555554, | |
| "grad_norm": 1.6410186290740967, | |
| "learning_rate": 2.7364438839848677e-05, | |
| "loss": 1.6826, | |
| "mean_token_accuracy": 0.6792753636837006, | |
| "num_tokens": 3306623.0, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 2.6293333333333333, | |
| "grad_norm": 1.4993571043014526, | |
| "learning_rate": 2.610340479192938e-05, | |
| "loss": 1.6629, | |
| "mean_token_accuracy": 0.6814461290836334, | |
| "num_tokens": 3329270.0, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 2.647111111111111, | |
| "grad_norm": 1.4495617151260376, | |
| "learning_rate": 2.484237074401009e-05, | |
| "loss": 1.6848, | |
| "mean_token_accuracy": 0.6769454509019852, | |
| "num_tokens": 3352533.0, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 2.664888888888889, | |
| "grad_norm": 1.5677741765975952, | |
| "learning_rate": 2.3581336696090794e-05, | |
| "loss": 1.6679, | |
| "mean_token_accuracy": 0.6842619329690933, | |
| "num_tokens": 3374132.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.6826666666666665, | |
| "grad_norm": 1.5430514812469482, | |
| "learning_rate": 2.23203026481715e-05, | |
| "loss": 1.7122, | |
| "mean_token_accuracy": 0.6701778277754784, | |
| "num_tokens": 3397176.0, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 2.7004444444444444, | |
| "grad_norm": 1.5498685836791992, | |
| "learning_rate": 2.1059268600252208e-05, | |
| "loss": 1.6631, | |
| "mean_token_accuracy": 0.6801098987460137, | |
| "num_tokens": 3418961.0, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 2.7182222222222223, | |
| "grad_norm": 1.5674185752868652, | |
| "learning_rate": 1.9798234552332915e-05, | |
| "loss": 1.6677, | |
| "mean_token_accuracy": 0.679096283018589, | |
| "num_tokens": 3441815.0, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 2.7359999999999998, | |
| "grad_norm": 1.4433151483535767, | |
| "learning_rate": 1.8537200504413622e-05, | |
| "loss": 1.6903, | |
| "mean_token_accuracy": 0.6747847631573677, | |
| "num_tokens": 3464582.0, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 2.7537777777777777, | |
| "grad_norm": 1.4974557161331177, | |
| "learning_rate": 1.7276166456494325e-05, | |
| "loss": 1.6533, | |
| "mean_token_accuracy": 0.683028981089592, | |
| "num_tokens": 3486694.0, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 2.7715555555555556, | |
| "grad_norm": 1.4934104681015015, | |
| "learning_rate": 1.6015132408575032e-05, | |
| "loss": 1.699, | |
| "mean_token_accuracy": 0.6773065477609634, | |
| "num_tokens": 3508908.0, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 2.7893333333333334, | |
| "grad_norm": 1.5283113718032837, | |
| "learning_rate": 1.4754098360655739e-05, | |
| "loss": 1.6329, | |
| "mean_token_accuracy": 0.6867676630616188, | |
| "num_tokens": 3530306.0, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 2.8071111111111113, | |
| "grad_norm": 1.6120567321777344, | |
| "learning_rate": 1.3493064312736444e-05, | |
| "loss": 1.7164, | |
| "mean_token_accuracy": 0.6712423786520958, | |
| "num_tokens": 3552880.0, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 2.824888888888889, | |
| "grad_norm": 1.5219435691833496, | |
| "learning_rate": 1.223203026481715e-05, | |
| "loss": 1.7018, | |
| "mean_token_accuracy": 0.6758274272084236, | |
| "num_tokens": 3575800.0, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 2.8426666666666667, | |
| "grad_norm": 1.5157614946365356, | |
| "learning_rate": 1.0970996216897856e-05, | |
| "loss": 1.6983, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.8426666666666667, | |
| "eval_loss": 1.9110984802246094, | |
| "eval_mean_token_accuracy": 0.6461525177955627, | |
| "eval_num_tokens": 3598214.0, | |
| "eval_runtime": 30.748, | |
| "eval_samples_per_second": 32.522, | |
| "eval_steps_per_second": 8.131, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 2.8604444444444446, | |
| "grad_norm": 1.5881015062332153, | |
| "learning_rate": 9.709962168978563e-06, | |
| "loss": 1.679, | |
| "mean_token_accuracy": 0.6803914837539196, | |
| "num_tokens": 3620353.0, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 2.878222222222222, | |
| "grad_norm": 1.5353198051452637, | |
| "learning_rate": 8.448928121059268e-06, | |
| "loss": 1.681, | |
| "mean_token_accuracy": 0.6778546258807182, | |
| "num_tokens": 3642281.0, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 2.896, | |
| "grad_norm": 1.5839284658432007, | |
| "learning_rate": 7.187894073139975e-06, | |
| "loss": 1.6397, | |
| "mean_token_accuracy": 0.6827977553009987, | |
| "num_tokens": 3664853.0, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 2.913777777777778, | |
| "grad_norm": 1.6904590129852295, | |
| "learning_rate": 5.926860025220681e-06, | |
| "loss": 1.7405, | |
| "mean_token_accuracy": 0.6691433653235436, | |
| "num_tokens": 3688123.0, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 2.9315555555555557, | |
| "grad_norm": 1.5607562065124512, | |
| "learning_rate": 4.665825977301387e-06, | |
| "loss": 1.7285, | |
| "mean_token_accuracy": 0.6671519264578819, | |
| "num_tokens": 3710934.0, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 2.9493333333333336, | |
| "grad_norm": 1.613752841949463, | |
| "learning_rate": 3.404791929382094e-06, | |
| "loss": 1.7251, | |
| "mean_token_accuracy": 0.676729716360569, | |
| "num_tokens": 3733298.0, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 2.967111111111111, | |
| "grad_norm": 1.4459993839263916, | |
| "learning_rate": 2.1437578814628e-06, | |
| "loss": 1.729, | |
| "mean_token_accuracy": 0.6713886946439743, | |
| "num_tokens": 3755641.0, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 2.984888888888889, | |
| "grad_norm": 1.6163533926010132, | |
| "learning_rate": 8.827238335435058e-07, | |
| "loss": 1.6633, | |
| "mean_token_accuracy": 0.6792998388409615, | |
| "num_tokens": 3777938.0, | |
| "step": 1680 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1686, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3348842111787008e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |