| { | |
| "best_metric": 1.9140400886535645, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Qwen2-7B_pct_default_r16/checkpoint-376", | |
| "epoch": 0.9996779388083736, | |
| "eval_steps": 8, | |
| "global_step": 388, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0025764895330112722, | |
| "grad_norm": 6.036927223205566, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.0463, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010305958132045089, | |
| "grad_norm": 4.005067825317383, | |
| "learning_rate": 5e-05, | |
| "loss": 2.1077, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020611916264090178, | |
| "grad_norm": 7.223879814147949, | |
| "learning_rate": 0.0001, | |
| "loss": 2.031, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020611916264090178, | |
| "eval_loss": 1.9518073797225952, | |
| "eval_runtime": 125.9043, | |
| "eval_samples_per_second": 1.946, | |
| "eval_steps_per_second": 0.977, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.030917874396135265, | |
| "grad_norm": 2.8582210540771484, | |
| "learning_rate": 9.997266286704631e-05, | |
| "loss": 2.0048, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.041223832528180356, | |
| "grad_norm": 2.455995559692383, | |
| "learning_rate": 9.989068136093873e-05, | |
| "loss": 2.0039, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.041223832528180356, | |
| "eval_loss": 1.9526941776275635, | |
| "eval_runtime": 107.8801, | |
| "eval_samples_per_second": 2.271, | |
| "eval_steps_per_second": 1.14, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05152979066022544, | |
| "grad_norm": 2.7170777320861816, | |
| "learning_rate": 9.975414512725057e-05, | |
| "loss": 1.9595, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06183574879227053, | |
| "grad_norm": 3.476964235305786, | |
| "learning_rate": 9.956320346634876e-05, | |
| "loss": 2.0258, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06183574879227053, | |
| "eval_loss": 1.9363595247268677, | |
| "eval_runtime": 123.7728, | |
| "eval_samples_per_second": 1.979, | |
| "eval_steps_per_second": 0.994, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07214170692431562, | |
| "grad_norm": 2.7470412254333496, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 2.0143, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08244766505636071, | |
| "grad_norm": 2.5656278133392334, | |
| "learning_rate": 9.901899829374047e-05, | |
| "loss": 1.9476, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08244766505636071, | |
| "eval_loss": 1.9395873546600342, | |
| "eval_runtime": 237.2101, | |
| "eval_samples_per_second": 1.033, | |
| "eval_steps_per_second": 0.519, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0927536231884058, | |
| "grad_norm": 2.428863763809204, | |
| "learning_rate": 9.86663298624003e-05, | |
| "loss": 2.0166, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10305958132045089, | |
| "grad_norm": 2.5703530311584473, | |
| "learning_rate": 9.826044551386744e-05, | |
| "loss": 1.993, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10305958132045089, | |
| "eval_loss": 1.9390852451324463, | |
| "eval_runtime": 125.0953, | |
| "eval_samples_per_second": 1.959, | |
| "eval_steps_per_second": 0.983, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11336553945249597, | |
| "grad_norm": 2.8227152824401855, | |
| "learning_rate": 9.780178907671789e-05, | |
| "loss": 1.9189, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12367149758454106, | |
| "grad_norm": 2.5152082443237305, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 1.9924, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12367149758454106, | |
| "eval_loss": 1.9412620067596436, | |
| "eval_runtime": 125.5452, | |
| "eval_samples_per_second": 1.951, | |
| "eval_steps_per_second": 0.98, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13397745571658615, | |
| "grad_norm": 2.8379340171813965, | |
| "learning_rate": 9.672822322997305e-05, | |
| "loss": 1.9297, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14428341384863125, | |
| "grad_norm": 2.782554864883423, | |
| "learning_rate": 9.611448774886924e-05, | |
| "loss": 2.0036, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14428341384863125, | |
| "eval_loss": 1.9407403469085693, | |
| "eval_runtime": 111.7215, | |
| "eval_samples_per_second": 2.193, | |
| "eval_steps_per_second": 1.101, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.15458937198067632, | |
| "grad_norm": 3.028449535369873, | |
| "learning_rate": 9.545032675245813e-05, | |
| "loss": 1.9739, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16489533011272142, | |
| "grad_norm": 3.124444007873535, | |
| "learning_rate": 9.473646649103818e-05, | |
| "loss": 1.9358, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16489533011272142, | |
| "eval_loss": 1.9378271102905273, | |
| "eval_runtime": 122.2691, | |
| "eval_samples_per_second": 2.004, | |
| "eval_steps_per_second": 1.006, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1752012882447665, | |
| "grad_norm": 2.85798716545105, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 2.0492, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.1855072463768116, | |
| "grad_norm": 2.860339403152466, | |
| "learning_rate": 9.316282404787871e-05, | |
| "loss": 1.9956, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1855072463768116, | |
| "eval_loss": 1.9400973320007324, | |
| "eval_runtime": 233.619, | |
| "eval_samples_per_second": 1.049, | |
| "eval_steps_per_second": 0.526, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19581320450885667, | |
| "grad_norm": 2.945218563079834, | |
| "learning_rate": 9.230476262104677e-05, | |
| "loss": 2.0209, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.20611916264090177, | |
| "grad_norm": 2.412891387939453, | |
| "learning_rate": 9.140044155740101e-05, | |
| "loss": 2.0183, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.20611916264090177, | |
| "eval_loss": 1.9399027824401855, | |
| "eval_runtime": 104.912, | |
| "eval_samples_per_second": 2.335, | |
| "eval_steps_per_second": 1.172, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21642512077294687, | |
| "grad_norm": 2.782628059387207, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.0703, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22673107890499195, | |
| "grad_norm": 2.9531919956207275, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 1.9952, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22673107890499195, | |
| "eval_loss": 1.9411203861236572, | |
| "eval_runtime": 125.6512, | |
| "eval_samples_per_second": 1.95, | |
| "eval_steps_per_second": 0.979, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 2.6295812129974365, | |
| "learning_rate": 8.842005554284296e-05, | |
| "loss": 1.916, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.24734299516908212, | |
| "grad_norm": 2.6426079273223877, | |
| "learning_rate": 8.73410738492077e-05, | |
| "loss": 1.9309, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.24734299516908212, | |
| "eval_loss": 1.941253662109375, | |
| "eval_runtime": 116.9383, | |
| "eval_samples_per_second": 2.095, | |
| "eval_steps_per_second": 1.052, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2576489533011272, | |
| "grad_norm": 2.405819892883301, | |
| "learning_rate": 8.622126023955446e-05, | |
| "loss": 2.0233, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2679549114331723, | |
| "grad_norm": 2.601391315460205, | |
| "learning_rate": 8.506183921362443e-05, | |
| "loss": 2.0042, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2679549114331723, | |
| "eval_loss": 1.9425873756408691, | |
| "eval_runtime": 109.1015, | |
| "eval_samples_per_second": 2.246, | |
| "eval_steps_per_second": 1.127, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2782608695652174, | |
| "grad_norm": 2.4709646701812744, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 2.0291, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.2885668276972625, | |
| "grad_norm": 2.8112270832061768, | |
| "learning_rate": 8.262928807620843e-05, | |
| "loss": 1.8885, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2885668276972625, | |
| "eval_loss": 1.9404640197753906, | |
| "eval_runtime": 249.4306, | |
| "eval_samples_per_second": 0.982, | |
| "eval_steps_per_second": 0.493, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.29887278582930754, | |
| "grad_norm": 2.589830160140991, | |
| "learning_rate": 8.135881792367686e-05, | |
| "loss": 2.0415, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.30917874396135264, | |
| "grad_norm": 2.830230712890625, | |
| "learning_rate": 8.005405736415126e-05, | |
| "loss": 1.9462, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.30917874396135264, | |
| "eval_loss": 1.9408910274505615, | |
| "eval_runtime": 118.2548, | |
| "eval_samples_per_second": 2.072, | |
| "eval_steps_per_second": 1.04, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.31948470209339774, | |
| "grad_norm": 3.2373831272125244, | |
| "learning_rate": 7.871643313414718e-05, | |
| "loss": 2.0322, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.32979066022544284, | |
| "grad_norm": 2.929025650024414, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 1.9787, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.32979066022544284, | |
| "eval_loss": 1.9440549612045288, | |
| "eval_runtime": 124.819, | |
| "eval_samples_per_second": 1.963, | |
| "eval_steps_per_second": 0.985, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.34009661835748795, | |
| "grad_norm": 3.1260762214660645, | |
| "learning_rate": 7.594847868906076e-05, | |
| "loss": 1.9843, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.350402576489533, | |
| "grad_norm": 3.138162851333618, | |
| "learning_rate": 7.452117519152542e-05, | |
| "loss": 1.9647, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.350402576489533, | |
| "eval_loss": 1.9407563209533691, | |
| "eval_runtime": 113.5508, | |
| "eval_samples_per_second": 2.158, | |
| "eval_steps_per_second": 1.083, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3607085346215781, | |
| "grad_norm": 2.764012336730957, | |
| "learning_rate": 7.30670581489344e-05, | |
| "loss": 1.9292, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3710144927536232, | |
| "grad_norm": 2.7180254459381104, | |
| "learning_rate": 7.158771761692464e-05, | |
| "loss": 1.9391, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.3710144927536232, | |
| "eval_loss": 1.9398295879364014, | |
| "eval_runtime": 208.0995, | |
| "eval_samples_per_second": 1.177, | |
| "eval_steps_per_second": 0.591, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.3813204508856683, | |
| "grad_norm": 2.800187826156616, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 1.9976, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.39162640901771334, | |
| "grad_norm": 2.7626631259918213, | |
| "learning_rate": 6.855986244591104e-05, | |
| "loss": 2.0038, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.39162640901771334, | |
| "eval_loss": 1.9389182329177856, | |
| "eval_runtime": 233.4305, | |
| "eval_samples_per_second": 1.05, | |
| "eval_steps_per_second": 0.527, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.40193236714975844, | |
| "grad_norm": 2.7909302711486816, | |
| "learning_rate": 6.701465872208216e-05, | |
| "loss": 1.9963, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.41223832528180354, | |
| "grad_norm": 2.133431911468506, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.0412, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.41223832528180354, | |
| "eval_loss": 1.9401731491088867, | |
| "eval_runtime": 120.3714, | |
| "eval_samples_per_second": 2.035, | |
| "eval_steps_per_second": 1.022, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.42254428341384864, | |
| "grad_norm": 2.4993631839752197, | |
| "learning_rate": 6.387014543809223e-05, | |
| "loss": 2.0386, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43285024154589374, | |
| "grad_norm": 2.903820753097534, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 2.0523, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43285024154589374, | |
| "eval_loss": 1.9371063709259033, | |
| "eval_runtime": 117.7362, | |
| "eval_samples_per_second": 2.081, | |
| "eval_steps_per_second": 1.045, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.4431561996779388, | |
| "grad_norm": 2.638428211212158, | |
| "learning_rate": 6.066498153718735e-05, | |
| "loss": 1.9451, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.4534621578099839, | |
| "grad_norm": 2.898460626602173, | |
| "learning_rate": 5.90440267166055e-05, | |
| "loss": 1.9364, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.4534621578099839, | |
| "eval_loss": 1.9393686056137085, | |
| "eval_runtime": 130.6388, | |
| "eval_samples_per_second": 1.875, | |
| "eval_steps_per_second": 0.942, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.463768115942029, | |
| "grad_norm": 3.116927146911621, | |
| "learning_rate": 5.74131823855921e-05, | |
| "loss": 2.0401, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 2.4770333766937256, | |
| "learning_rate": 5.577423184847932e-05, | |
| "loss": 1.9805, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "eval_loss": 1.9394539594650269, | |
| "eval_runtime": 221.6098, | |
| "eval_samples_per_second": 1.106, | |
| "eval_steps_per_second": 0.555, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.48438003220611914, | |
| "grad_norm": 2.262666702270508, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 1.9786, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.49468599033816424, | |
| "grad_norm": 2.686546802520752, | |
| "learning_rate": 5.247918773366112e-05, | |
| "loss": 1.9935, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.49468599033816424, | |
| "eval_loss": 1.9380260705947876, | |
| "eval_runtime": 121.9702, | |
| "eval_samples_per_second": 2.009, | |
| "eval_steps_per_second": 1.008, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5049919484702093, | |
| "grad_norm": 2.820659875869751, | |
| "learning_rate": 5.0826697238317935e-05, | |
| "loss": 1.9155, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5152979066022544, | |
| "grad_norm": 2.929722785949707, | |
| "learning_rate": 4.917330276168208e-05, | |
| "loss": 1.9342, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5152979066022544, | |
| "eval_loss": 1.9345794916152954, | |
| "eval_runtime": 131.2895, | |
| "eval_samples_per_second": 1.866, | |
| "eval_steps_per_second": 0.937, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5256038647342995, | |
| "grad_norm": 2.6336963176727295, | |
| "learning_rate": 4.7520812266338885e-05, | |
| "loss": 2.0489, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5359098228663446, | |
| "grad_norm": 2.551440477371216, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 1.9708, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5359098228663446, | |
| "eval_loss": 1.9360543489456177, | |
| "eval_runtime": 119.4716, | |
| "eval_samples_per_second": 2.051, | |
| "eval_steps_per_second": 1.03, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5462157809983897, | |
| "grad_norm": 2.525174856185913, | |
| "learning_rate": 4.4225768151520694e-05, | |
| "loss": 1.9851, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "grad_norm": 2.6649765968322754, | |
| "learning_rate": 4.2586817614407895e-05, | |
| "loss": 2.0128, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "eval_loss": 1.9354972839355469, | |
| "eval_runtime": 128.1483, | |
| "eval_samples_per_second": 1.912, | |
| "eval_steps_per_second": 0.96, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5668276972624798, | |
| "grad_norm": 2.538203001022339, | |
| "learning_rate": 4.095597328339452e-05, | |
| "loss": 2.0375, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.577133655394525, | |
| "grad_norm": 2.639700412750244, | |
| "learning_rate": 3.933501846281267e-05, | |
| "loss": 1.9416, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.577133655394525, | |
| "eval_loss": 1.930440902709961, | |
| "eval_runtime": 219.1333, | |
| "eval_samples_per_second": 1.118, | |
| "eval_steps_per_second": 0.561, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.58743961352657, | |
| "grad_norm": 2.561579704284668, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 2.022, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5977455716586151, | |
| "grad_norm": 2.42655611038208, | |
| "learning_rate": 3.612985456190778e-05, | |
| "loss": 1.9658, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5977455716586151, | |
| "eval_loss": 1.9349409341812134, | |
| "eval_runtime": 107.6562, | |
| "eval_samples_per_second": 2.276, | |
| "eval_steps_per_second": 1.143, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6080515297906602, | |
| "grad_norm": 2.426743268966675, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.9736, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.6183574879227053, | |
| "grad_norm": 2.663156747817993, | |
| "learning_rate": 3.298534127791785e-05, | |
| "loss": 1.9161, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6183574879227053, | |
| "eval_loss": 1.9257853031158447, | |
| "eval_runtime": 50.4859, | |
| "eval_samples_per_second": 4.853, | |
| "eval_steps_per_second": 2.436, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6286634460547504, | |
| "grad_norm": 2.280381679534912, | |
| "learning_rate": 3.144013755408895e-05, | |
| "loss": 1.9576, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6389694041867955, | |
| "grad_norm": 2.5414199829101562, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 1.94, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6389694041867955, | |
| "eval_loss": 1.9257922172546387, | |
| "eval_runtime": 128.998, | |
| "eval_samples_per_second": 1.899, | |
| "eval_steps_per_second": 0.954, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6492753623188405, | |
| "grad_norm": 2.5111727714538574, | |
| "learning_rate": 2.8412282383075363e-05, | |
| "loss": 1.9636, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6595813204508857, | |
| "grad_norm": 2.4180471897125244, | |
| "learning_rate": 2.693294185106562e-05, | |
| "loss": 1.9908, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6595813204508857, | |
| "eval_loss": 1.9244165420532227, | |
| "eval_runtime": 124.6471, | |
| "eval_samples_per_second": 1.966, | |
| "eval_steps_per_second": 0.987, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6698872785829307, | |
| "grad_norm": 2.398568868637085, | |
| "learning_rate": 2.547882480847461e-05, | |
| "loss": 1.9779, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6801932367149759, | |
| "grad_norm": 2.521932601928711, | |
| "learning_rate": 2.405152131093926e-05, | |
| "loss": 1.9169, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6801932367149759, | |
| "eval_loss": 1.9241654872894287, | |
| "eval_runtime": 232.9073, | |
| "eval_samples_per_second": 1.052, | |
| "eval_steps_per_second": 0.528, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6904991948470209, | |
| "grad_norm": 2.5789191722869873, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 2.015, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.700805152979066, | |
| "grad_norm": 2.303483724594116, | |
| "learning_rate": 2.128356686585282e-05, | |
| "loss": 1.9868, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.700805152979066, | |
| "eval_loss": 1.921614408493042, | |
| "eval_runtime": 108.603, | |
| "eval_samples_per_second": 2.256, | |
| "eval_steps_per_second": 1.133, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 2.581958293914795, | |
| "learning_rate": 1.9945942635848748e-05, | |
| "loss": 1.9239, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7214170692431562, | |
| "grad_norm": 2.9368388652801514, | |
| "learning_rate": 1.8641182076323148e-05, | |
| "loss": 1.8737, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7214170692431562, | |
| "eval_loss": 1.9208799600601196, | |
| "eval_runtime": 123.5168, | |
| "eval_samples_per_second": 1.984, | |
| "eval_steps_per_second": 0.996, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7317230273752013, | |
| "grad_norm": 2.706059455871582, | |
| "learning_rate": 1.7370711923791567e-05, | |
| "loss": 1.9982, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7420289855072464, | |
| "grad_norm": 2.5310757160186768, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 2.0166, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7420289855072464, | |
| "eval_loss": 1.919832706451416, | |
| "eval_runtime": 124.8133, | |
| "eval_samples_per_second": 1.963, | |
| "eval_steps_per_second": 0.985, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7523349436392914, | |
| "grad_norm": 2.52554988861084, | |
| "learning_rate": 1.4938160786375572e-05, | |
| "loss": 1.9849, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7626409017713366, | |
| "grad_norm": 2.3502135276794434, | |
| "learning_rate": 1.3778739760445552e-05, | |
| "loss": 1.9246, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7626409017713366, | |
| "eval_loss": 1.918765664100647, | |
| "eval_runtime": 100.865, | |
| "eval_samples_per_second": 2.429, | |
| "eval_steps_per_second": 1.219, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7729468599033816, | |
| "grad_norm": 2.43375301361084, | |
| "learning_rate": 1.2658926150792322e-05, | |
| "loss": 1.9995, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7832528180354267, | |
| "grad_norm": 2.5884594917297363, | |
| "learning_rate": 1.157994445715706e-05, | |
| "loss": 1.9418, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7832528180354267, | |
| "eval_loss": 1.9197624921798706, | |
| "eval_runtime": 223.502, | |
| "eval_samples_per_second": 1.096, | |
| "eval_steps_per_second": 0.55, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7935587761674718, | |
| "grad_norm": 2.3900303840637207, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 1.9024, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.8038647342995169, | |
| "grad_norm": 2.4108939170837402, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.9417, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8038647342995169, | |
| "eval_loss": 1.9172273874282837, | |
| "eval_runtime": 126.9286, | |
| "eval_samples_per_second": 1.93, | |
| "eval_steps_per_second": 0.969, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.814170692431562, | |
| "grad_norm": 2.105642557144165, | |
| "learning_rate": 8.599558442598998e-06, | |
| "loss": 1.9766, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8244766505636071, | |
| "grad_norm": 2.292396306991577, | |
| "learning_rate": 7.695237378953223e-06, | |
| "loss": 1.9652, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8244766505636071, | |
| "eval_loss": 1.9169241189956665, | |
| "eval_runtime": 118.9904, | |
| "eval_samples_per_second": 2.059, | |
| "eval_steps_per_second": 1.034, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8347826086956521, | |
| "grad_norm": 2.122368574142456, | |
| "learning_rate": 6.837175952121306e-06, | |
| "loss": 2.004, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8450885668276973, | |
| "grad_norm": 2.2409677505493164, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 1.9715, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8450885668276973, | |
| "eval_loss": 1.9170624017715454, | |
| "eval_runtime": 111.3633, | |
| "eval_samples_per_second": 2.2, | |
| "eval_steps_per_second": 1.104, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8553945249597423, | |
| "grad_norm": 2.5156409740448, | |
| "learning_rate": 5.263533508961827e-06, | |
| "loss": 1.9188, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8657004830917875, | |
| "grad_norm": 2.1870505809783936, | |
| "learning_rate": 4.549673247541875e-06, | |
| "loss": 1.9634, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8657004830917875, | |
| "eval_loss": 1.9158947467803955, | |
| "eval_runtime": 234.5094, | |
| "eval_samples_per_second": 1.045, | |
| "eval_steps_per_second": 0.524, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8760064412238325, | |
| "grad_norm": 2.428206443786621, | |
| "learning_rate": 3.885512251130763e-06, | |
| "loss": 1.9557, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8863123993558776, | |
| "grad_norm": 2.5798287391662598, | |
| "learning_rate": 3.271776770026963e-06, | |
| "loss": 1.9566, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8863123993558776, | |
| "eval_loss": 1.9153064489364624, | |
| "eval_runtime": 139.4736, | |
| "eval_samples_per_second": 1.757, | |
| "eval_steps_per_second": 0.882, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8966183574879227, | |
| "grad_norm": 2.345715045928955, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 1.8952, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9069243156199678, | |
| "grad_norm": 2.8093199729919434, | |
| "learning_rate": 2.1982109232821178e-06, | |
| "loss": 1.9277, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9069243156199678, | |
| "eval_loss": 1.9147199392318726, | |
| "eval_runtime": 148.1878, | |
| "eval_samples_per_second": 1.653, | |
| "eval_steps_per_second": 0.83, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9172302737520129, | |
| "grad_norm": 2.609008312225342, | |
| "learning_rate": 1.7395544861325718e-06, | |
| "loss": 2.0425, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.927536231884058, | |
| "grad_norm": 2.2126529216766357, | |
| "learning_rate": 1.333670137599713e-06, | |
| "loss": 2.0087, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.927536231884058, | |
| "eval_loss": 1.9142426252365112, | |
| "eval_runtime": 129.4444, | |
| "eval_samples_per_second": 1.893, | |
| "eval_steps_per_second": 0.95, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.937842190016103, | |
| "grad_norm": 2.5125489234924316, | |
| "learning_rate": 9.810017062595322e-07, | |
| "loss": 1.9817, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 2.638725996017456, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 1.921, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "eval_loss": 1.9142839908599854, | |
| "eval_runtime": 131.7251, | |
| "eval_samples_per_second": 1.86, | |
| "eval_steps_per_second": 0.934, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9584541062801932, | |
| "grad_norm": 2.400455951690674, | |
| "learning_rate": 4.367965336512403e-07, | |
| "loss": 1.8983, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9687600644122383, | |
| "grad_norm": 2.5474154949188232, | |
| "learning_rate": 2.458548727494292e-07, | |
| "loss": 1.9842, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9687600644122383, | |
| "eval_loss": 1.9140400886535645, | |
| "eval_runtime": 251.3639, | |
| "eval_samples_per_second": 0.975, | |
| "eval_steps_per_second": 0.489, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9790660225442834, | |
| "grad_norm": 2.566486120223999, | |
| "learning_rate": 1.0931863906127327e-07, | |
| "loss": 1.8793, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9893719806763285, | |
| "grad_norm": 2.8591597080230713, | |
| "learning_rate": 2.7337132953697554e-08, | |
| "loss": 1.8825, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9893719806763285, | |
| "eval_loss": 1.914077639579773, | |
| "eval_runtime": 110.406, | |
| "eval_samples_per_second": 2.219, | |
| "eval_steps_per_second": 1.114, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9996779388083736, | |
| "grad_norm": 2.4471967220306396, | |
| "learning_rate": 0.0, | |
| "loss": 1.9747, | |
| "step": 388 | |
| } | |
| ], | |
| "logging_steps": 4, | |
| "max_steps": 388, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.427475676604334e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |