| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 189, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.016, | |
| "grad_norm": 0.8398123979568481, | |
| "learning_rate": 0.0, | |
| "loss": 4.0858, | |
| "mean_token_accuracy": 0.34444408491253853, | |
| "num_tokens": 2645.0, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.032, | |
| "grad_norm": 0.7567061185836792, | |
| "learning_rate": 1.0526315789473685e-06, | |
| "loss": 3.9664, | |
| "mean_token_accuracy": 0.3604096956551075, | |
| "num_tokens": 5533.0, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.048, | |
| "grad_norm": 0.7942607998847961, | |
| "learning_rate": 2.105263157894737e-06, | |
| "loss": 3.9819, | |
| "mean_token_accuracy": 0.36218464002013206, | |
| "num_tokens": 8279.0, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.064, | |
| "grad_norm": 0.8740888237953186, | |
| "learning_rate": 3.157894736842105e-06, | |
| "loss": 4.1531, | |
| "mean_token_accuracy": 0.3526028133928776, | |
| "num_tokens": 10817.0, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 0.7643747329711914, | |
| "learning_rate": 4.210526315789474e-06, | |
| "loss": 3.998, | |
| "mean_token_accuracy": 0.3644312396645546, | |
| "num_tokens": 13463.0, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.096, | |
| "grad_norm": 0.8161706328392029, | |
| "learning_rate": 5.263157894736842e-06, | |
| "loss": 4.0672, | |
| "mean_token_accuracy": 0.3655700273811817, | |
| "num_tokens": 16041.0, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.112, | |
| "grad_norm": 0.8353070020675659, | |
| "learning_rate": 6.31578947368421e-06, | |
| "loss": 3.9997, | |
| "mean_token_accuracy": 0.3524913415312767, | |
| "num_tokens": 18708.0, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.128, | |
| "grad_norm": 0.761987566947937, | |
| "learning_rate": 7.368421052631579e-06, | |
| "loss": 3.9241, | |
| "mean_token_accuracy": 0.3674253560602665, | |
| "num_tokens": 21349.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.144, | |
| "grad_norm": 0.7558256983757019, | |
| "learning_rate": 8.421052631578948e-06, | |
| "loss": 4.0323, | |
| "mean_token_accuracy": 0.36221769079566, | |
| "num_tokens": 24265.0, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.7886160016059875, | |
| "learning_rate": 9.473684210526315e-06, | |
| "loss": 4.0076, | |
| "mean_token_accuracy": 0.3525189645588398, | |
| "num_tokens": 26782.0, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.176, | |
| "grad_norm": 0.8333842158317566, | |
| "learning_rate": 1.0526315789473684e-05, | |
| "loss": 3.9085, | |
| "mean_token_accuracy": 0.36505160480737686, | |
| "num_tokens": 29261.0, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.192, | |
| "grad_norm": 0.7002362012863159, | |
| "learning_rate": 1.1578947368421053e-05, | |
| "loss": 3.85, | |
| "mean_token_accuracy": 0.37056851387023926, | |
| "num_tokens": 32061.0, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.208, | |
| "grad_norm": 0.7784165143966675, | |
| "learning_rate": 1.263157894736842e-05, | |
| "loss": 3.968, | |
| "mean_token_accuracy": 0.35847294703125954, | |
| "num_tokens": 34787.0, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.224, | |
| "grad_norm": 0.7894275784492493, | |
| "learning_rate": 1.3684210526315791e-05, | |
| "loss": 3.9739, | |
| "mean_token_accuracy": 0.36424775049090385, | |
| "num_tokens": 37402.0, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.7131537199020386, | |
| "learning_rate": 1.4736842105263159e-05, | |
| "loss": 3.8961, | |
| "mean_token_accuracy": 0.35714147984981537, | |
| "num_tokens": 40147.0, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.256, | |
| "grad_norm": 0.6734439134597778, | |
| "learning_rate": 1.578947368421053e-05, | |
| "loss": 3.8699, | |
| "mean_token_accuracy": 0.3641946502029896, | |
| "num_tokens": 43086.0, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.272, | |
| "grad_norm": 0.7871038317680359, | |
| "learning_rate": 1.6842105263157896e-05, | |
| "loss": 3.9493, | |
| "mean_token_accuracy": 0.35068848729133606, | |
| "num_tokens": 45876.0, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.288, | |
| "grad_norm": 0.6782894730567932, | |
| "learning_rate": 1.7894736842105264e-05, | |
| "loss": 3.8261, | |
| "mean_token_accuracy": 0.3632819801568985, | |
| "num_tokens": 48598.0, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.304, | |
| "grad_norm": 0.8898190259933472, | |
| "learning_rate": 1.894736842105263e-05, | |
| "loss": 4.1267, | |
| "mean_token_accuracy": 0.34038742259144783, | |
| "num_tokens": 51214.0, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.811371922492981, | |
| "learning_rate": 2e-05, | |
| "loss": 3.9657, | |
| "mean_token_accuracy": 0.355673860758543, | |
| "num_tokens": 53848.0, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.336, | |
| "grad_norm": 0.8319784998893738, | |
| "learning_rate": 1.988235294117647e-05, | |
| "loss": 3.9473, | |
| "mean_token_accuracy": 0.3650202564895153, | |
| "num_tokens": 56275.0, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.352, | |
| "grad_norm": 0.8142621517181396, | |
| "learning_rate": 1.9764705882352945e-05, | |
| "loss": 4.0495, | |
| "mean_token_accuracy": 0.34513410925865173, | |
| "num_tokens": 58775.0, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.368, | |
| "grad_norm": 0.8011351227760315, | |
| "learning_rate": 1.964705882352941e-05, | |
| "loss": 3.915, | |
| "mean_token_accuracy": 0.3376794904470444, | |
| "num_tokens": 61460.0, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.384, | |
| "grad_norm": 0.7196938395500183, | |
| "learning_rate": 1.9529411764705885e-05, | |
| "loss": 3.7644, | |
| "mean_token_accuracy": 0.36175281181931496, | |
| "num_tokens": 64184.0, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 0.7823134660720825, | |
| "learning_rate": 1.9411764705882355e-05, | |
| "loss": 3.8629, | |
| "mean_token_accuracy": 0.357443917542696, | |
| "num_tokens": 67001.0, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.416, | |
| "grad_norm": 0.7930059432983398, | |
| "learning_rate": 1.9294117647058825e-05, | |
| "loss": 3.7996, | |
| "mean_token_accuracy": 0.35901443287730217, | |
| "num_tokens": 69645.0, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.432, | |
| "grad_norm": 0.8171699047088623, | |
| "learning_rate": 1.9176470588235298e-05, | |
| "loss": 3.8514, | |
| "mean_token_accuracy": 0.36160001531243324, | |
| "num_tokens": 72226.0, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.448, | |
| "grad_norm": 0.7647044062614441, | |
| "learning_rate": 1.9058823529411764e-05, | |
| "loss": 3.8502, | |
| "mean_token_accuracy": 0.3531816340982914, | |
| "num_tokens": 75040.0, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.464, | |
| "grad_norm": 0.7744542360305786, | |
| "learning_rate": 1.8941176470588238e-05, | |
| "loss": 3.6849, | |
| "mean_token_accuracy": 0.3711949326097965, | |
| "num_tokens": 77771.0, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.770905077457428, | |
| "learning_rate": 1.8823529411764708e-05, | |
| "loss": 3.6916, | |
| "mean_token_accuracy": 0.36217380687594414, | |
| "num_tokens": 80540.0, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.496, | |
| "grad_norm": 0.7673084139823914, | |
| "learning_rate": 1.8705882352941178e-05, | |
| "loss": 3.7276, | |
| "mean_token_accuracy": 0.3703504465520382, | |
| "num_tokens": 83105.0, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.512, | |
| "grad_norm": 0.7939592599868774, | |
| "learning_rate": 1.8588235294117647e-05, | |
| "loss": 3.7805, | |
| "mean_token_accuracy": 0.3589933402836323, | |
| "num_tokens": 85857.0, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.528, | |
| "grad_norm": 0.8425813913345337, | |
| "learning_rate": 1.847058823529412e-05, | |
| "loss": 3.6672, | |
| "mean_token_accuracy": 0.3700985908508301, | |
| "num_tokens": 88677.0, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.544, | |
| "grad_norm": 0.7484491467475891, | |
| "learning_rate": 1.8352941176470587e-05, | |
| "loss": 3.6443, | |
| "mean_token_accuracy": 0.37437862157821655, | |
| "num_tokens": 91435.0, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.7103927731513977, | |
| "learning_rate": 1.823529411764706e-05, | |
| "loss": 3.6768, | |
| "mean_token_accuracy": 0.3736225366592407, | |
| "num_tokens": 94208.0, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.576, | |
| "grad_norm": 0.7449000477790833, | |
| "learning_rate": 1.811764705882353e-05, | |
| "loss": 3.7151, | |
| "mean_token_accuracy": 0.36662693321704865, | |
| "num_tokens": 96770.0, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.592, | |
| "grad_norm": 0.6955702900886536, | |
| "learning_rate": 1.8e-05, | |
| "loss": 3.6739, | |
| "mean_token_accuracy": 0.3633931763470173, | |
| "num_tokens": 99672.0, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.608, | |
| "grad_norm": 0.8391516208648682, | |
| "learning_rate": 1.7882352941176474e-05, | |
| "loss": 3.7544, | |
| "mean_token_accuracy": 0.36131446436047554, | |
| "num_tokens": 102394.0, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.624, | |
| "grad_norm": 0.6383936405181885, | |
| "learning_rate": 1.776470588235294e-05, | |
| "loss": 3.4877, | |
| "mean_token_accuracy": 0.38343239575624466, | |
| "num_tokens": 105125.0, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.7163877487182617, | |
| "learning_rate": 1.7647058823529414e-05, | |
| "loss": 3.5337, | |
| "mean_token_accuracy": 0.39500385895371437, | |
| "num_tokens": 107651.0, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.656, | |
| "grad_norm": 0.7929803729057312, | |
| "learning_rate": 1.7529411764705884e-05, | |
| "loss": 3.5737, | |
| "mean_token_accuracy": 0.3653530888259411, | |
| "num_tokens": 110246.0, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.672, | |
| "grad_norm": 0.7210948467254639, | |
| "learning_rate": 1.7411764705882353e-05, | |
| "loss": 3.6813, | |
| "mean_token_accuracy": 0.3572908975183964, | |
| "num_tokens": 112860.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.688, | |
| "grad_norm": 0.7377284169197083, | |
| "learning_rate": 1.7294117647058827e-05, | |
| "loss": 3.622, | |
| "mean_token_accuracy": 0.3657442480325699, | |
| "num_tokens": 115779.0, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.704, | |
| "grad_norm": 0.703372061252594, | |
| "learning_rate": 1.7176470588235293e-05, | |
| "loss": 3.6187, | |
| "mean_token_accuracy": 0.3651924841105938, | |
| "num_tokens": 118421.0, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 0.6265890598297119, | |
| "learning_rate": 1.7058823529411767e-05, | |
| "loss": 3.581, | |
| "mean_token_accuracy": 0.36536476761102676, | |
| "num_tokens": 121364.0, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.736, | |
| "grad_norm": 0.6594799757003784, | |
| "learning_rate": 1.6941176470588237e-05, | |
| "loss": 3.5259, | |
| "mean_token_accuracy": 0.35779330506920815, | |
| "num_tokens": 124309.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.752, | |
| "grad_norm": 0.6088898777961731, | |
| "learning_rate": 1.6823529411764706e-05, | |
| "loss": 3.4377, | |
| "mean_token_accuracy": 0.38217825815081596, | |
| "num_tokens": 127054.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.768, | |
| "grad_norm": 0.6634587645530701, | |
| "learning_rate": 1.670588235294118e-05, | |
| "loss": 3.5712, | |
| "mean_token_accuracy": 0.3662005178630352, | |
| "num_tokens": 129686.0, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.784, | |
| "grad_norm": 0.646621584892273, | |
| "learning_rate": 1.658823529411765e-05, | |
| "loss": 3.5078, | |
| "mean_token_accuracy": 0.3694671206176281, | |
| "num_tokens": 132429.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.6244218945503235, | |
| "learning_rate": 1.647058823529412e-05, | |
| "loss": 3.4824, | |
| "mean_token_accuracy": 0.3822427801787853, | |
| "num_tokens": 135256.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.816, | |
| "grad_norm": 0.5937113165855408, | |
| "learning_rate": 1.635294117647059e-05, | |
| "loss": 3.4575, | |
| "mean_token_accuracy": 0.3822938874363899, | |
| "num_tokens": 137969.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.832, | |
| "grad_norm": 0.5072384476661682, | |
| "learning_rate": 1.623529411764706e-05, | |
| "loss": 3.3456, | |
| "mean_token_accuracy": 0.38250361382961273, | |
| "num_tokens": 141031.0, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.848, | |
| "grad_norm": 0.5952928066253662, | |
| "learning_rate": 1.6117647058823533e-05, | |
| "loss": 3.4753, | |
| "mean_token_accuracy": 0.392064705491066, | |
| "num_tokens": 143652.0, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.864, | |
| "grad_norm": 0.5355834364891052, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 3.4376, | |
| "mean_token_accuracy": 0.3916851729154587, | |
| "num_tokens": 146480.0, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 0.5446826219558716, | |
| "learning_rate": 1.5882352941176473e-05, | |
| "loss": 3.478, | |
| "mean_token_accuracy": 0.3886304572224617, | |
| "num_tokens": 149041.0, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.896, | |
| "grad_norm": 0.5369603037834167, | |
| "learning_rate": 1.5764705882352943e-05, | |
| "loss": 3.4426, | |
| "mean_token_accuracy": 0.3753341734409332, | |
| "num_tokens": 151713.0, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.912, | |
| "grad_norm": 0.49587222933769226, | |
| "learning_rate": 1.5647058823529412e-05, | |
| "loss": 3.395, | |
| "mean_token_accuracy": 0.39359384775161743, | |
| "num_tokens": 154568.0, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.928, | |
| "grad_norm": 0.5416087508201599, | |
| "learning_rate": 1.5529411764705882e-05, | |
| "loss": 3.4106, | |
| "mean_token_accuracy": 0.38531995937228203, | |
| "num_tokens": 157239.0, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.944, | |
| "grad_norm": 0.5001004934310913, | |
| "learning_rate": 1.5411764705882356e-05, | |
| "loss": 3.2466, | |
| "mean_token_accuracy": 0.4112909585237503, | |
| "num_tokens": 160044.0, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 0.5069928765296936, | |
| "learning_rate": 1.5294117647058822e-05, | |
| "loss": 3.5089, | |
| "mean_token_accuracy": 0.37355804070830345, | |
| "num_tokens": 162656.0, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.976, | |
| "grad_norm": 0.5107595324516296, | |
| "learning_rate": 1.5176470588235295e-05, | |
| "loss": 3.4407, | |
| "mean_token_accuracy": 0.3820053040981293, | |
| "num_tokens": 165250.0, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.992, | |
| "grad_norm": 0.4934955835342407, | |
| "learning_rate": 1.5058823529411765e-05, | |
| "loss": 3.4864, | |
| "mean_token_accuracy": 0.38048501312732697, | |
| "num_tokens": 167981.0, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.49148648977279663, | |
| "learning_rate": 1.4941176470588237e-05, | |
| "loss": 3.3985, | |
| "mean_token_accuracy": 0.37495024502277374, | |
| "num_tokens": 169409.0, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.016, | |
| "grad_norm": 0.45961520075798035, | |
| "learning_rate": 1.4823529411764707e-05, | |
| "loss": 3.3384, | |
| "mean_token_accuracy": 0.382406834512949, | |
| "num_tokens": 172136.0, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.032, | |
| "grad_norm": 0.45267099142074585, | |
| "learning_rate": 1.4705882352941179e-05, | |
| "loss": 3.248, | |
| "mean_token_accuracy": 0.4042637459933758, | |
| "num_tokens": 174726.0, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.048, | |
| "grad_norm": 0.4088224470615387, | |
| "learning_rate": 1.4588235294117647e-05, | |
| "loss": 3.3626, | |
| "mean_token_accuracy": 0.3834773004055023, | |
| "num_tokens": 177667.0, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.064, | |
| "grad_norm": 0.449268639087677, | |
| "learning_rate": 1.4470588235294118e-05, | |
| "loss": 3.3041, | |
| "mean_token_accuracy": 0.398575808852911, | |
| "num_tokens": 180362.0, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 0.4171251058578491, | |
| "learning_rate": 1.435294117647059e-05, | |
| "loss": 3.2265, | |
| "mean_token_accuracy": 0.4078996740281582, | |
| "num_tokens": 183195.0, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.096, | |
| "grad_norm": 0.4535180330276489, | |
| "learning_rate": 1.423529411764706e-05, | |
| "loss": 3.3778, | |
| "mean_token_accuracy": 0.39422746002674103, | |
| "num_tokens": 185582.0, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.112, | |
| "grad_norm": 0.42611923813819885, | |
| "learning_rate": 1.4117647058823532e-05, | |
| "loss": 3.3583, | |
| "mean_token_accuracy": 0.3894058130681515, | |
| "num_tokens": 188027.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1280000000000001, | |
| "grad_norm": 0.4249477684497833, | |
| "learning_rate": 1.4e-05, | |
| "loss": 3.2751, | |
| "mean_token_accuracy": 0.390879075974226, | |
| "num_tokens": 190623.0, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 1.144, | |
| "grad_norm": 0.3975493013858795, | |
| "learning_rate": 1.3882352941176471e-05, | |
| "loss": 3.339, | |
| "mean_token_accuracy": 0.38816238567233086, | |
| "num_tokens": 193449.0, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 0.42475345730781555, | |
| "learning_rate": 1.3764705882352943e-05, | |
| "loss": 3.2581, | |
| "mean_token_accuracy": 0.3975643031299114, | |
| "num_tokens": 196055.0, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 1.176, | |
| "grad_norm": 0.3927607238292694, | |
| "learning_rate": 1.3647058823529413e-05, | |
| "loss": 3.2769, | |
| "mean_token_accuracy": 0.39931680262088776, | |
| "num_tokens": 198819.0, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 1.192, | |
| "grad_norm": 0.40998390316963196, | |
| "learning_rate": 1.3529411764705885e-05, | |
| "loss": 3.2761, | |
| "mean_token_accuracy": 0.39620402827858925, | |
| "num_tokens": 201629.0, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.208, | |
| "grad_norm": 0.4474492073059082, | |
| "learning_rate": 1.3411764705882353e-05, | |
| "loss": 3.3408, | |
| "mean_token_accuracy": 0.38205572962760925, | |
| "num_tokens": 204201.0, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 1.224, | |
| "grad_norm": 0.3776940107345581, | |
| "learning_rate": 1.3294117647058824e-05, | |
| "loss": 3.1786, | |
| "mean_token_accuracy": 0.39778972417116165, | |
| "num_tokens": 207089.0, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.3904842436313629, | |
| "learning_rate": 1.3176470588235294e-05, | |
| "loss": 3.285, | |
| "mean_token_accuracy": 0.4057367295026779, | |
| "num_tokens": 209870.0, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.256, | |
| "grad_norm": 0.4232347011566162, | |
| "learning_rate": 1.3058823529411766e-05, | |
| "loss": 3.2816, | |
| "mean_token_accuracy": 0.40294618532061577, | |
| "num_tokens": 212347.0, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.272, | |
| "grad_norm": 0.38529646396636963, | |
| "learning_rate": 1.2941176470588238e-05, | |
| "loss": 3.2211, | |
| "mean_token_accuracy": 0.4018060527741909, | |
| "num_tokens": 215177.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.288, | |
| "grad_norm": 0.40373918414115906, | |
| "learning_rate": 1.2823529411764707e-05, | |
| "loss": 3.2446, | |
| "mean_token_accuracy": 0.4000708982348442, | |
| "num_tokens": 217938.0, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.304, | |
| "grad_norm": 0.3896755874156952, | |
| "learning_rate": 1.2705882352941177e-05, | |
| "loss": 3.2614, | |
| "mean_token_accuracy": 0.3885551504790783, | |
| "num_tokens": 220654.0, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 0.4156249463558197, | |
| "learning_rate": 1.2588235294117647e-05, | |
| "loss": 3.2249, | |
| "mean_token_accuracy": 0.394759364426136, | |
| "num_tokens": 223250.0, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 1.336, | |
| "grad_norm": 0.3963974416255951, | |
| "learning_rate": 1.2470588235294119e-05, | |
| "loss": 3.2153, | |
| "mean_token_accuracy": 0.4060470461845398, | |
| "num_tokens": 226119.0, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.3519999999999999, | |
| "grad_norm": 0.41384029388427734, | |
| "learning_rate": 1.235294117647059e-05, | |
| "loss": 3.2743, | |
| "mean_token_accuracy": 0.3922521695494652, | |
| "num_tokens": 228571.0, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.3679999999999999, | |
| "grad_norm": 0.41883495450019836, | |
| "learning_rate": 1.223529411764706e-05, | |
| "loss": 3.2732, | |
| "mean_token_accuracy": 0.4045484773814678, | |
| "num_tokens": 231100.0, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.384, | |
| "grad_norm": 0.3822110891342163, | |
| "learning_rate": 1.211764705882353e-05, | |
| "loss": 3.2107, | |
| "mean_token_accuracy": 0.4009360149502754, | |
| "num_tokens": 233918.0, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.3723452091217041, | |
| "learning_rate": 1.2e-05, | |
| "loss": 3.2225, | |
| "mean_token_accuracy": 0.3909096457064152, | |
| "num_tokens": 236828.0, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 1.416, | |
| "grad_norm": 0.37734636664390564, | |
| "learning_rate": 1.1882352941176472e-05, | |
| "loss": 3.3024, | |
| "mean_token_accuracy": 0.38941896334290504, | |
| "num_tokens": 239546.0, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 1.432, | |
| "grad_norm": 0.3807719647884369, | |
| "learning_rate": 1.1764705882352942e-05, | |
| "loss": 3.2143, | |
| "mean_token_accuracy": 0.40611620619893074, | |
| "num_tokens": 242254.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.448, | |
| "grad_norm": 0.3992786407470703, | |
| "learning_rate": 1.1647058823529413e-05, | |
| "loss": 3.2275, | |
| "mean_token_accuracy": 0.40825914219021797, | |
| "num_tokens": 244878.0, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 1.464, | |
| "grad_norm": 0.3822246491909027, | |
| "learning_rate": 1.1529411764705882e-05, | |
| "loss": 3.1385, | |
| "mean_token_accuracy": 0.40679260343313217, | |
| "num_tokens": 247604.0, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 0.397044837474823, | |
| "learning_rate": 1.1411764705882353e-05, | |
| "loss": 3.1844, | |
| "mean_token_accuracy": 0.39359012246131897, | |
| "num_tokens": 250362.0, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.496, | |
| "grad_norm": 0.38244199752807617, | |
| "learning_rate": 1.1294117647058825e-05, | |
| "loss": 3.1378, | |
| "mean_token_accuracy": 0.41278792172670364, | |
| "num_tokens": 253123.0, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 1.512, | |
| "grad_norm": 0.389908105134964, | |
| "learning_rate": 1.1176470588235295e-05, | |
| "loss": 3.2156, | |
| "mean_token_accuracy": 0.3887946680188179, | |
| "num_tokens": 255761.0, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.528, | |
| "grad_norm": 0.3996700346469879, | |
| "learning_rate": 1.1058823529411766e-05, | |
| "loss": 3.1115, | |
| "mean_token_accuracy": 0.4130321890115738, | |
| "num_tokens": 258418.0, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.544, | |
| "grad_norm": 0.40736863017082214, | |
| "learning_rate": 1.0941176470588238e-05, | |
| "loss": 3.2651, | |
| "mean_token_accuracy": 0.3938223719596863, | |
| "num_tokens": 261043.0, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 0.3875584304332733, | |
| "learning_rate": 1.0823529411764706e-05, | |
| "loss": 3.1533, | |
| "mean_token_accuracy": 0.39770323410630226, | |
| "num_tokens": 263871.0, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 1.576, | |
| "grad_norm": 0.38282302021980286, | |
| "learning_rate": 1.0705882352941178e-05, | |
| "loss": 3.178, | |
| "mean_token_accuracy": 0.3948086127638817, | |
| "num_tokens": 266722.0, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.592, | |
| "grad_norm": 0.3955148458480835, | |
| "learning_rate": 1.0588235294117648e-05, | |
| "loss": 3.0313, | |
| "mean_token_accuracy": 0.41876740753650665, | |
| "num_tokens": 269425.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.608, | |
| "grad_norm": 0.39854276180267334, | |
| "learning_rate": 1.047058823529412e-05, | |
| "loss": 3.1829, | |
| "mean_token_accuracy": 0.3948264420032501, | |
| "num_tokens": 272089.0, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 1.624, | |
| "grad_norm": 0.4018220901489258, | |
| "learning_rate": 1.035294117647059e-05, | |
| "loss": 3.1655, | |
| "mean_token_accuracy": 0.40457021445035934, | |
| "num_tokens": 274752.0, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 1.6400000000000001, | |
| "grad_norm": 0.37287989258766174, | |
| "learning_rate": 1.023529411764706e-05, | |
| "loss": 3.0649, | |
| "mean_token_accuracy": 0.40372075513005257, | |
| "num_tokens": 277484.0, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 1.6560000000000001, | |
| "grad_norm": 0.3763083517551422, | |
| "learning_rate": 1.011764705882353e-05, | |
| "loss": 3.019, | |
| "mean_token_accuracy": 0.4213622435927391, | |
| "num_tokens": 280400.0, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 1.6720000000000002, | |
| "grad_norm": 0.4037889242172241, | |
| "learning_rate": 1e-05, | |
| "loss": 3.1122, | |
| "mean_token_accuracy": 0.41004806756973267, | |
| "num_tokens": 283006.0, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.688, | |
| "grad_norm": 0.3816189169883728, | |
| "learning_rate": 9.882352941176472e-06, | |
| "loss": 3.1132, | |
| "mean_token_accuracy": 0.4011266306042671, | |
| "num_tokens": 285691.0, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 1.704, | |
| "grad_norm": 0.38289642333984375, | |
| "learning_rate": 9.764705882352942e-06, | |
| "loss": 3.069, | |
| "mean_token_accuracy": 0.4087080657482147, | |
| "num_tokens": 288406.0, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 0.3701270520687103, | |
| "learning_rate": 9.647058823529412e-06, | |
| "loss": 3.0733, | |
| "mean_token_accuracy": 0.4002223238348961, | |
| "num_tokens": 291219.0, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 1.736, | |
| "grad_norm": 0.3989318311214447, | |
| "learning_rate": 9.529411764705882e-06, | |
| "loss": 3.1161, | |
| "mean_token_accuracy": 0.4087665304541588, | |
| "num_tokens": 293951.0, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 1.752, | |
| "grad_norm": 0.385897696018219, | |
| "learning_rate": 9.411764705882354e-06, | |
| "loss": 3.1857, | |
| "mean_token_accuracy": 0.40533048287034035, | |
| "num_tokens": 296709.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.768, | |
| "grad_norm": 0.38372376561164856, | |
| "learning_rate": 9.294117647058824e-06, | |
| "loss": 3.0504, | |
| "mean_token_accuracy": 0.40413976460695267, | |
| "num_tokens": 299503.0, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 1.784, | |
| "grad_norm": 0.396903395652771, | |
| "learning_rate": 9.176470588235294e-06, | |
| "loss": 3.0204, | |
| "mean_token_accuracy": 0.42540061101317406, | |
| "num_tokens": 302028.0, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.3853081464767456, | |
| "learning_rate": 9.058823529411765e-06, | |
| "loss": 3.0607, | |
| "mean_token_accuracy": 0.4026433639228344, | |
| "num_tokens": 304679.0, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 1.8159999999999998, | |
| "grad_norm": 0.384064644575119, | |
| "learning_rate": 8.941176470588237e-06, | |
| "loss": 3.0478, | |
| "mean_token_accuracy": 0.4026435390114784, | |
| "num_tokens": 307365.0, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.8319999999999999, | |
| "grad_norm": 0.38211265206336975, | |
| "learning_rate": 8.823529411764707e-06, | |
| "loss": 3.0388, | |
| "mean_token_accuracy": 0.40072883665561676, | |
| "num_tokens": 310209.0, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.8479999999999999, | |
| "grad_norm": 0.3584171533584595, | |
| "learning_rate": 8.705882352941177e-06, | |
| "loss": 2.9238, | |
| "mean_token_accuracy": 0.4217926971614361, | |
| "num_tokens": 313159.0, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 1.8639999999999999, | |
| "grad_norm": 0.3544124662876129, | |
| "learning_rate": 8.588235294117647e-06, | |
| "loss": 2.9881, | |
| "mean_token_accuracy": 0.40666651353240013, | |
| "num_tokens": 316146.0, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.41010335087776184, | |
| "learning_rate": 8.470588235294118e-06, | |
| "loss": 3.0058, | |
| "mean_token_accuracy": 0.4097953774034977, | |
| "num_tokens": 318616.0, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 1.896, | |
| "grad_norm": 0.3789406418800354, | |
| "learning_rate": 8.35294117647059e-06, | |
| "loss": 2.9718, | |
| "mean_token_accuracy": 0.41611189767718315, | |
| "num_tokens": 321456.0, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 1.912, | |
| "grad_norm": 0.39727213978767395, | |
| "learning_rate": 8.23529411764706e-06, | |
| "loss": 3.0936, | |
| "mean_token_accuracy": 0.40342430397868156, | |
| "num_tokens": 324091.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.928, | |
| "grad_norm": 0.4349411725997925, | |
| "learning_rate": 8.11764705882353e-06, | |
| "loss": 3.0792, | |
| "mean_token_accuracy": 0.39731455221772194, | |
| "num_tokens": 326612.0, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 1.944, | |
| "grad_norm": 0.40542107820510864, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 3.0563, | |
| "mean_token_accuracy": 0.3993385322391987, | |
| "num_tokens": 329261.0, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 0.39871078729629517, | |
| "learning_rate": 7.882352941176471e-06, | |
| "loss": 3.1843, | |
| "mean_token_accuracy": 0.4031324125826359, | |
| "num_tokens": 331885.0, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 1.976, | |
| "grad_norm": 0.4009590148925781, | |
| "learning_rate": 7.764705882352941e-06, | |
| "loss": 3.0735, | |
| "mean_token_accuracy": 0.40051887556910515, | |
| "num_tokens": 334439.0, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 1.992, | |
| "grad_norm": 0.3671548366546631, | |
| "learning_rate": 7.647058823529411e-06, | |
| "loss": 3.0195, | |
| "mean_token_accuracy": 0.4079283997416496, | |
| "num_tokens": 337458.0, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.41459402441978455, | |
| "learning_rate": 7.529411764705883e-06, | |
| "loss": 3.021, | |
| "mean_token_accuracy": 0.3909439966082573, | |
| "num_tokens": 338818.0, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 2.016, | |
| "grad_norm": 0.35790422558784485, | |
| "learning_rate": 7.4117647058823535e-06, | |
| "loss": 2.9934, | |
| "mean_token_accuracy": 0.4115968458354473, | |
| "num_tokens": 341659.0, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 2.032, | |
| "grad_norm": 0.4216051399707794, | |
| "learning_rate": 7.294117647058823e-06, | |
| "loss": 3.0538, | |
| "mean_token_accuracy": 0.40418078750371933, | |
| "num_tokens": 344312.0, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 2.048, | |
| "grad_norm": 0.37006205320358276, | |
| "learning_rate": 7.176470588235295e-06, | |
| "loss": 3.029, | |
| "mean_token_accuracy": 0.39688674733042717, | |
| "num_tokens": 347118.0, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 2.064, | |
| "grad_norm": 0.3772274851799011, | |
| "learning_rate": 7.058823529411766e-06, | |
| "loss": 2.889, | |
| "mean_token_accuracy": 0.4363688565790653, | |
| "num_tokens": 349868.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 0.3636110723018646, | |
| "learning_rate": 6.941176470588236e-06, | |
| "loss": 2.9298, | |
| "mean_token_accuracy": 0.41682084277272224, | |
| "num_tokens": 352657.0, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 2.096, | |
| "grad_norm": 0.40421512722969055, | |
| "learning_rate": 6.8235294117647065e-06, | |
| "loss": 3.041, | |
| "mean_token_accuracy": 0.4117591343820095, | |
| "num_tokens": 355223.0, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 2.112, | |
| "grad_norm": 0.3686157166957855, | |
| "learning_rate": 6.705882352941176e-06, | |
| "loss": 3.106, | |
| "mean_token_accuracy": 0.3990732170641422, | |
| "num_tokens": 357978.0, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 2.128, | |
| "grad_norm": 0.36047035455703735, | |
| "learning_rate": 6.588235294117647e-06, | |
| "loss": 3.0028, | |
| "mean_token_accuracy": 0.40924155339598656, | |
| "num_tokens": 360823.0, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 2.144, | |
| "grad_norm": 0.39206644892692566, | |
| "learning_rate": 6.470588235294119e-06, | |
| "loss": 2.9553, | |
| "mean_token_accuracy": 0.4173494316637516, | |
| "num_tokens": 363404.0, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 0.3853868544101715, | |
| "learning_rate": 6.352941176470589e-06, | |
| "loss": 2.9031, | |
| "mean_token_accuracy": 0.4202713146805763, | |
| "num_tokens": 366095.0, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 2.176, | |
| "grad_norm": 0.37982177734375, | |
| "learning_rate": 6.2352941176470595e-06, | |
| "loss": 2.9148, | |
| "mean_token_accuracy": 0.41410698369145393, | |
| "num_tokens": 368762.0, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 2.192, | |
| "grad_norm": 0.378548264503479, | |
| "learning_rate": 6.11764705882353e-06, | |
| "loss": 2.8896, | |
| "mean_token_accuracy": 0.4176674447953701, | |
| "num_tokens": 371352.0, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 2.208, | |
| "grad_norm": 0.367887943983078, | |
| "learning_rate": 6e-06, | |
| "loss": 2.9974, | |
| "mean_token_accuracy": 0.41198988631367683, | |
| "num_tokens": 374161.0, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 2.224, | |
| "grad_norm": 0.39015528559684753, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 2.9768, | |
| "mean_token_accuracy": 0.41146713495254517, | |
| "num_tokens": 376784.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 0.3850547671318054, | |
| "learning_rate": 5.764705882352941e-06, | |
| "loss": 2.9659, | |
| "mean_token_accuracy": 0.412378154695034, | |
| "num_tokens": 379354.0, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 2.2560000000000002, | |
| "grad_norm": 0.4373947083950043, | |
| "learning_rate": 5.6470588235294125e-06, | |
| "loss": 3.069, | |
| "mean_token_accuracy": 0.41074345260858536, | |
| "num_tokens": 381690.0, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 2.2720000000000002, | |
| "grad_norm": 0.3996342718601227, | |
| "learning_rate": 5.529411764705883e-06, | |
| "loss": 3.014, | |
| "mean_token_accuracy": 0.40605925023555756, | |
| "num_tokens": 384266.0, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 2.288, | |
| "grad_norm": 0.38876214623451233, | |
| "learning_rate": 5.411764705882353e-06, | |
| "loss": 3.075, | |
| "mean_token_accuracy": 0.40350398793816566, | |
| "num_tokens": 386931.0, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 2.304, | |
| "grad_norm": 0.3649292290210724, | |
| "learning_rate": 5.294117647058824e-06, | |
| "loss": 2.9637, | |
| "mean_token_accuracy": 0.4175785072147846, | |
| "num_tokens": 389690.0, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 0.36741501092910767, | |
| "learning_rate": 5.176470588235295e-06, | |
| "loss": 2.8984, | |
| "mean_token_accuracy": 0.4225447066128254, | |
| "num_tokens": 392512.0, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 2.336, | |
| "grad_norm": 0.37910395860671997, | |
| "learning_rate": 5.058823529411765e-06, | |
| "loss": 2.9528, | |
| "mean_token_accuracy": 0.40937231108546257, | |
| "num_tokens": 395194.0, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 2.352, | |
| "grad_norm": 0.3879609704017639, | |
| "learning_rate": 4.941176470588236e-06, | |
| "loss": 3.0491, | |
| "mean_token_accuracy": 0.39538025483489037, | |
| "num_tokens": 397718.0, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 2.368, | |
| "grad_norm": 0.39440643787384033, | |
| "learning_rate": 4.823529411764706e-06, | |
| "loss": 2.9381, | |
| "mean_token_accuracy": 0.4080496206879616, | |
| "num_tokens": 400393.0, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 2.384, | |
| "grad_norm": 0.3876059353351593, | |
| "learning_rate": 4.705882352941177e-06, | |
| "loss": 3.0132, | |
| "mean_token_accuracy": 0.4006643071770668, | |
| "num_tokens": 403053.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 0.3781592547893524, | |
| "learning_rate": 4.588235294117647e-06, | |
| "loss": 3.009, | |
| "mean_token_accuracy": 0.40124019235372543, | |
| "num_tokens": 405694.0, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 2.416, | |
| "grad_norm": 0.3604719042778015, | |
| "learning_rate": 4.4705882352941184e-06, | |
| "loss": 2.9941, | |
| "mean_token_accuracy": 0.4007309637963772, | |
| "num_tokens": 408444.0, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 2.432, | |
| "grad_norm": 0.3836471438407898, | |
| "learning_rate": 4.352941176470588e-06, | |
| "loss": 2.9924, | |
| "mean_token_accuracy": 0.3976939432322979, | |
| "num_tokens": 411141.0, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 2.448, | |
| "grad_norm": 0.3532804548740387, | |
| "learning_rate": 4.235294117647059e-06, | |
| "loss": 2.9098, | |
| "mean_token_accuracy": 0.4064694568514824, | |
| "num_tokens": 414051.0, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 2.464, | |
| "grad_norm": 0.403434157371521, | |
| "learning_rate": 4.11764705882353e-06, | |
| "loss": 2.9763, | |
| "mean_token_accuracy": 0.40338218584656715, | |
| "num_tokens": 416708.0, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 0.3774099051952362, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.9725, | |
| "mean_token_accuracy": 0.41620006039738655, | |
| "num_tokens": 419347.0, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 2.496, | |
| "grad_norm": 0.35273030400276184, | |
| "learning_rate": 3.882352941176471e-06, | |
| "loss": 2.9244, | |
| "mean_token_accuracy": 0.42492661997675896, | |
| "num_tokens": 422064.0, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 2.512, | |
| "grad_norm": 0.398325651884079, | |
| "learning_rate": 3.7647058823529414e-06, | |
| "loss": 3.0626, | |
| "mean_token_accuracy": 0.40975887328386307, | |
| "num_tokens": 424614.0, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 2.528, | |
| "grad_norm": 0.35992905497550964, | |
| "learning_rate": 3.6470588235294117e-06, | |
| "loss": 2.9204, | |
| "mean_token_accuracy": 0.42144904658198357, | |
| "num_tokens": 427372.0, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 2.544, | |
| "grad_norm": 0.3729715347290039, | |
| "learning_rate": 3.529411764705883e-06, | |
| "loss": 2.9345, | |
| "mean_token_accuracy": 0.4180946573615074, | |
| "num_tokens": 430030.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 0.3701145350933075, | |
| "learning_rate": 3.4117647058823532e-06, | |
| "loss": 2.9872, | |
| "mean_token_accuracy": 0.4019898883998394, | |
| "num_tokens": 432855.0, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 2.576, | |
| "grad_norm": 0.341402530670166, | |
| "learning_rate": 3.2941176470588236e-06, | |
| "loss": 2.9675, | |
| "mean_token_accuracy": 0.405661266297102, | |
| "num_tokens": 435904.0, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 2.592, | |
| "grad_norm": 0.3595600128173828, | |
| "learning_rate": 3.1764705882352943e-06, | |
| "loss": 2.9506, | |
| "mean_token_accuracy": 0.421790212392807, | |
| "num_tokens": 438712.0, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 2.608, | |
| "grad_norm": 0.3787381947040558, | |
| "learning_rate": 3.058823529411765e-06, | |
| "loss": 3.0634, | |
| "mean_token_accuracy": 0.39874693378806114, | |
| "num_tokens": 441416.0, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 2.624, | |
| "grad_norm": 0.36589527130126953, | |
| "learning_rate": 2.9411764705882355e-06, | |
| "loss": 2.9687, | |
| "mean_token_accuracy": 0.40731603279709816, | |
| "num_tokens": 444167.0, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 0.3544527292251587, | |
| "learning_rate": 2.8235294117647062e-06, | |
| "loss": 2.8222, | |
| "mean_token_accuracy": 0.4249437265098095, | |
| "num_tokens": 446982.0, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 2.656, | |
| "grad_norm": 0.39410585165023804, | |
| "learning_rate": 2.7058823529411766e-06, | |
| "loss": 2.8953, | |
| "mean_token_accuracy": 0.42447903379797935, | |
| "num_tokens": 449436.0, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 2.672, | |
| "grad_norm": 0.3637503981590271, | |
| "learning_rate": 2.5882352941176473e-06, | |
| "loss": 2.9117, | |
| "mean_token_accuracy": 0.4190538227558136, | |
| "num_tokens": 452176.0, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 2.6879999999999997, | |
| "grad_norm": 0.3520418703556061, | |
| "learning_rate": 2.470588235294118e-06, | |
| "loss": 2.9634, | |
| "mean_token_accuracy": 0.4060833677649498, | |
| "num_tokens": 454997.0, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 2.7039999999999997, | |
| "grad_norm": 0.3733985722064972, | |
| "learning_rate": 2.3529411764705885e-06, | |
| "loss": 2.982, | |
| "mean_token_accuracy": 0.4063185714185238, | |
| "num_tokens": 457616.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.7199999999999998, | |
| "grad_norm": 0.33130913972854614, | |
| "learning_rate": 2.2352941176470592e-06, | |
| "loss": 2.8688, | |
| "mean_token_accuracy": 0.4284527339041233, | |
| "num_tokens": 460652.0, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 2.7359999999999998, | |
| "grad_norm": 0.36535510420799255, | |
| "learning_rate": 2.1176470588235296e-06, | |
| "loss": 2.8824, | |
| "mean_token_accuracy": 0.4171196520328522, | |
| "num_tokens": 463439.0, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.752, | |
| "grad_norm": 0.38491347432136536, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.917, | |
| "mean_token_accuracy": 0.4062299467623234, | |
| "num_tokens": 466079.0, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 2.768, | |
| "grad_norm": 0.3783172070980072, | |
| "learning_rate": 1.8823529411764707e-06, | |
| "loss": 2.9191, | |
| "mean_token_accuracy": 0.41218286380171776, | |
| "num_tokens": 468713.0, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 2.784, | |
| "grad_norm": 0.3626379668712616, | |
| "learning_rate": 1.7647058823529414e-06, | |
| "loss": 2.8956, | |
| "mean_token_accuracy": 0.4100930877029896, | |
| "num_tokens": 471530.0, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.38149014115333557, | |
| "learning_rate": 1.6470588235294118e-06, | |
| "loss": 2.9867, | |
| "mean_token_accuracy": 0.400991253554821, | |
| "num_tokens": 474170.0, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 2.816, | |
| "grad_norm": 0.3559909462928772, | |
| "learning_rate": 1.5294117647058826e-06, | |
| "loss": 2.8443, | |
| "mean_token_accuracy": 0.4292943999171257, | |
| "num_tokens": 477004.0, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 2.832, | |
| "grad_norm": 0.3727106750011444, | |
| "learning_rate": 1.4117647058823531e-06, | |
| "loss": 2.8494, | |
| "mean_token_accuracy": 0.42553965747356415, | |
| "num_tokens": 479614.0, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 2.848, | |
| "grad_norm": 0.34643471240997314, | |
| "learning_rate": 1.2941176470588237e-06, | |
| "loss": 2.947, | |
| "mean_token_accuracy": 0.4042917937040329, | |
| "num_tokens": 482633.0, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 2.864, | |
| "grad_norm": 0.3776470124721527, | |
| "learning_rate": 1.1764705882352942e-06, | |
| "loss": 2.9869, | |
| "mean_token_accuracy": 0.4118799567222595, | |
| "num_tokens": 485248.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 0.35878056287765503, | |
| "learning_rate": 1.0588235294117648e-06, | |
| "loss": 2.9014, | |
| "mean_token_accuracy": 0.4140179418027401, | |
| "num_tokens": 488082.0, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 2.896, | |
| "grad_norm": 0.3923113942146301, | |
| "learning_rate": 9.411764705882353e-07, | |
| "loss": 2.9649, | |
| "mean_token_accuracy": 0.4091031290590763, | |
| "num_tokens": 490791.0, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 2.912, | |
| "grad_norm": 0.36566361784935, | |
| "learning_rate": 8.235294117647059e-07, | |
| "loss": 2.8308, | |
| "mean_token_accuracy": 0.40432415530085564, | |
| "num_tokens": 493562.0, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 2.928, | |
| "grad_norm": 0.3739163875579834, | |
| "learning_rate": 7.058823529411766e-07, | |
| "loss": 2.9332, | |
| "mean_token_accuracy": 0.415198165923357, | |
| "num_tokens": 496296.0, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 2.944, | |
| "grad_norm": 0.35979267954826355, | |
| "learning_rate": 5.882352941176471e-07, | |
| "loss": 2.8786, | |
| "mean_token_accuracy": 0.41218165680766106, | |
| "num_tokens": 499079.0, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.37098458409309387, | |
| "learning_rate": 4.7058823529411767e-07, | |
| "loss": 2.8579, | |
| "mean_token_accuracy": 0.4170387051999569, | |
| "num_tokens": 501660.0, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 2.976, | |
| "grad_norm": 0.3667754530906677, | |
| "learning_rate": 3.529411764705883e-07, | |
| "loss": 2.9582, | |
| "mean_token_accuracy": 0.4101652540266514, | |
| "num_tokens": 504498.0, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 2.992, | |
| "grad_norm": 0.41497012972831726, | |
| "learning_rate": 2.3529411764705883e-07, | |
| "loss": 2.9947, | |
| "mean_token_accuracy": 0.4088564068078995, | |
| "num_tokens": 506810.0, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.4135372042655945, | |
| "learning_rate": 1.1764705882352942e-07, | |
| "loss": 2.9861, | |
| "mean_token_accuracy": 0.4060775488615036, | |
| "num_tokens": 508227.0, | |
| "step": 189 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 189, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1087844825174016.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |