| { | |
| "best_metric": 1.9137802124023438, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Qwen2-7B_pct_reverse_r16/checkpoint-368", | |
| "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": 6.036957740783691, | |
| "learning_rate": 1.25e-05, | |
| "loss": 2.0463, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010305958132045089, | |
| "grad_norm": 4.019509315490723, | |
| "learning_rate": 5e-05, | |
| "loss": 2.1076, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020611916264090178, | |
| "grad_norm": 10.257355690002441, | |
| "learning_rate": 0.0001, | |
| "loss": 2.0341, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020611916264090178, | |
| "eval_loss": 1.9528307914733887, | |
| "eval_runtime": 45.8734, | |
| "eval_samples_per_second": 5.341, | |
| "eval_steps_per_second": 2.681, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.030917874396135265, | |
| "grad_norm": 2.879976272583008, | |
| "learning_rate": 9.997266286704631e-05, | |
| "loss": 2.0054, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.041223832528180356, | |
| "grad_norm": 2.654456615447998, | |
| "learning_rate": 9.989068136093873e-05, | |
| "loss": 2.0078, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.041223832528180356, | |
| "eval_loss": 1.9550031423568726, | |
| "eval_runtime": 44.7469, | |
| "eval_samples_per_second": 5.475, | |
| "eval_steps_per_second": 2.749, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05152979066022544, | |
| "grad_norm": 2.678391695022583, | |
| "learning_rate": 9.975414512725057e-05, | |
| "loss": 1.9626, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06183574879227053, | |
| "grad_norm": 3.189049005508423, | |
| "learning_rate": 9.956320346634876e-05, | |
| "loss": 2.0269, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06183574879227053, | |
| "eval_loss": 1.9356740713119507, | |
| "eval_runtime": 45.2011, | |
| "eval_samples_per_second": 5.42, | |
| "eval_steps_per_second": 2.721, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07214170692431562, | |
| "grad_norm": 2.709564447402954, | |
| "learning_rate": 9.931806517013612e-05, | |
| "loss": 2.0135, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08244766505636071, | |
| "grad_norm": 2.585526466369629, | |
| "learning_rate": 9.901899829374047e-05, | |
| "loss": 1.9472, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08244766505636071, | |
| "eval_loss": 1.9405081272125244, | |
| "eval_runtime": 81.9576, | |
| "eval_samples_per_second": 2.989, | |
| "eval_steps_per_second": 1.501, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0927536231884058, | |
| "grad_norm": 2.4268321990966797, | |
| "learning_rate": 9.86663298624003e-05, | |
| "loss": 2.0166, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10305958132045089, | |
| "grad_norm": 2.597174882888794, | |
| "learning_rate": 9.826044551386744e-05, | |
| "loss": 1.993, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10305958132045089, | |
| "eval_loss": 1.9380648136138916, | |
| "eval_runtime": 45.2717, | |
| "eval_samples_per_second": 5.412, | |
| "eval_steps_per_second": 2.717, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11336553945249597, | |
| "grad_norm": 2.8009908199310303, | |
| "learning_rate": 9.780178907671789e-05, | |
| "loss": 1.9178, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12367149758454106, | |
| "grad_norm": 2.5525169372558594, | |
| "learning_rate": 9.729086208503174e-05, | |
| "loss": 1.9936, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12367149758454106, | |
| "eval_loss": 1.9402122497558594, | |
| "eval_runtime": 43.9726, | |
| "eval_samples_per_second": 5.572, | |
| "eval_steps_per_second": 2.797, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13397745571658615, | |
| "grad_norm": 2.791377305984497, | |
| "learning_rate": 9.672822322997305e-05, | |
| "loss": 1.9291, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14428341384863125, | |
| "grad_norm": 2.8142764568328857, | |
| "learning_rate": 9.611448774886924e-05, | |
| "loss": 2.0043, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14428341384863125, | |
| "eval_loss": 1.9410146474838257, | |
| "eval_runtime": 44.3537, | |
| "eval_samples_per_second": 5.524, | |
| "eval_steps_per_second": 2.773, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.15458937198067632, | |
| "grad_norm": 3.0118305683135986, | |
| "learning_rate": 9.545032675245813e-05, | |
| "loss": 1.9752, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16489533011272142, | |
| "grad_norm": 3.247464656829834, | |
| "learning_rate": 9.473646649103818e-05, | |
| "loss": 1.9356, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16489533011272142, | |
| "eval_loss": 1.9368629455566406, | |
| "eval_runtime": 44.9235, | |
| "eval_samples_per_second": 5.454, | |
| "eval_steps_per_second": 2.738, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1752012882447665, | |
| "grad_norm": 2.7969186305999756, | |
| "learning_rate": 9.397368756032445e-05, | |
| "loss": 2.0485, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.1855072463768116, | |
| "grad_norm": 2.8225085735321045, | |
| "learning_rate": 9.316282404787871e-05, | |
| "loss": 1.9953, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1855072463768116, | |
| "eval_loss": 1.9395699501037598, | |
| "eval_runtime": 77.6661, | |
| "eval_samples_per_second": 3.155, | |
| "eval_steps_per_second": 1.584, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19581320450885667, | |
| "grad_norm": 2.9584035873413086, | |
| "learning_rate": 9.230476262104677e-05, | |
| "loss": 2.0198, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.20611916264090177, | |
| "grad_norm": 2.4394288063049316, | |
| "learning_rate": 9.140044155740101e-05, | |
| "loss": 2.0184, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.20611916264090177, | |
| "eval_loss": 1.940533995628357, | |
| "eval_runtime": 43.9748, | |
| "eval_samples_per_second": 5.571, | |
| "eval_steps_per_second": 2.797, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21642512077294687, | |
| "grad_norm": 2.9622068405151367, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.0708, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22673107890499195, | |
| "grad_norm": 3.059290647506714, | |
| "learning_rate": 8.945702546981969e-05, | |
| "loss": 1.995, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22673107890499195, | |
| "eval_loss": 1.940983772277832, | |
| "eval_runtime": 44.7809, | |
| "eval_samples_per_second": 5.471, | |
| "eval_steps_per_second": 2.747, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 2.5659518241882324, | |
| "learning_rate": 8.842005554284296e-05, | |
| "loss": 1.9141, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.24734299516908212, | |
| "grad_norm": 2.770142078399658, | |
| "learning_rate": 8.73410738492077e-05, | |
| "loss": 1.9307, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.24734299516908212, | |
| "eval_loss": 1.9407052993774414, | |
| "eval_runtime": 44.3926, | |
| "eval_samples_per_second": 5.519, | |
| "eval_steps_per_second": 2.771, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2576489533011272, | |
| "grad_norm": 2.419646978378296, | |
| "learning_rate": 8.622126023955446e-05, | |
| "loss": 2.0221, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2679549114331723, | |
| "grad_norm": 2.4743905067443848, | |
| "learning_rate": 8.506183921362443e-05, | |
| "loss": 2.0037, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2679549114331723, | |
| "eval_loss": 1.9413758516311646, | |
| "eval_runtime": 43.7595, | |
| "eval_samples_per_second": 5.599, | |
| "eval_steps_per_second": 2.811, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2782608695652174, | |
| "grad_norm": 2.4670968055725098, | |
| "learning_rate": 8.386407858128706e-05, | |
| "loss": 2.0279, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.2885668276972625, | |
| "grad_norm": 2.8095762729644775, | |
| "learning_rate": 8.262928807620843e-05, | |
| "loss": 1.889, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2885668276972625, | |
| "eval_loss": 1.939664363861084, | |
| "eval_runtime": 79.1358, | |
| "eval_samples_per_second": 3.096, | |
| "eval_steps_per_second": 1.554, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.29887278582930754, | |
| "grad_norm": 2.5172698497772217, | |
| "learning_rate": 8.135881792367686e-05, | |
| "loss": 2.0408, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.30917874396135264, | |
| "grad_norm": 2.727034330368042, | |
| "learning_rate": 8.005405736415126e-05, | |
| "loss": 1.9455, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.30917874396135264, | |
| "eval_loss": 1.9401391744613647, | |
| "eval_runtime": 44.1305, | |
| "eval_samples_per_second": 5.552, | |
| "eval_steps_per_second": 2.787, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.31948470209339774, | |
| "grad_norm": 2.9831223487854004, | |
| "learning_rate": 7.871643313414718e-05, | |
| "loss": 2.032, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.32979066022544284, | |
| "grad_norm": 2.715611457824707, | |
| "learning_rate": 7.734740790612136e-05, | |
| "loss": 1.9789, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.32979066022544284, | |
| "eval_loss": 1.9438152313232422, | |
| "eval_runtime": 43.6559, | |
| "eval_samples_per_second": 5.612, | |
| "eval_steps_per_second": 2.817, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.34009661835748795, | |
| "grad_norm": 2.9471704959869385, | |
| "learning_rate": 7.594847868906076e-05, | |
| "loss": 1.9841, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.350402576489533, | |
| "grad_norm": 2.823962926864624, | |
| "learning_rate": 7.452117519152542e-05, | |
| "loss": 1.9642, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.350402576489533, | |
| "eval_loss": 1.940843939781189, | |
| "eval_runtime": 45.2787, | |
| "eval_samples_per_second": 5.411, | |
| "eval_steps_per_second": 2.717, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3607085346215781, | |
| "grad_norm": 2.5701866149902344, | |
| "learning_rate": 7.30670581489344e-05, | |
| "loss": 1.928, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3710144927536232, | |
| "grad_norm": 2.5628817081451416, | |
| "learning_rate": 7.158771761692464e-05, | |
| "loss": 1.9387, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.3710144927536232, | |
| "eval_loss": 1.9405008554458618, | |
| "eval_runtime": 77.1208, | |
| "eval_samples_per_second": 3.177, | |
| "eval_steps_per_second": 1.595, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.3813204508856683, | |
| "grad_norm": 2.6887667179107666, | |
| "learning_rate": 7.008477123264848e-05, | |
| "loss": 1.9983, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.39162640901771334, | |
| "grad_norm": 2.5755767822265625, | |
| "learning_rate": 6.855986244591104e-05, | |
| "loss": 2.0036, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.39162640901771334, | |
| "eval_loss": 1.9393625259399414, | |
| "eval_runtime": 77.0654, | |
| "eval_samples_per_second": 3.179, | |
| "eval_steps_per_second": 1.596, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.40193236714975844, | |
| "grad_norm": 2.772606134414673, | |
| "learning_rate": 6.701465872208216e-05, | |
| "loss": 1.9969, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.41223832528180354, | |
| "grad_norm": 2.1044411659240723, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.0407, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.41223832528180354, | |
| "eval_loss": 1.9393056631088257, | |
| "eval_runtime": 44.7521, | |
| "eval_samples_per_second": 5.475, | |
| "eval_steps_per_second": 2.748, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.42254428341384864, | |
| "grad_norm": 2.5281455516815186, | |
| "learning_rate": 6.387014543809223e-05, | |
| "loss": 2.0383, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43285024154589374, | |
| "grad_norm": 2.8692080974578857, | |
| "learning_rate": 6.227427435703997e-05, | |
| "loss": 2.0519, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43285024154589374, | |
| "eval_loss": 1.9384679794311523, | |
| "eval_runtime": 44.1246, | |
| "eval_samples_per_second": 5.552, | |
| "eval_steps_per_second": 2.788, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.4431561996779388, | |
| "grad_norm": 2.7248945236206055, | |
| "learning_rate": 6.066498153718735e-05, | |
| "loss": 1.9455, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.4534621578099839, | |
| "grad_norm": 2.897691011428833, | |
| "learning_rate": 5.90440267166055e-05, | |
| "loss": 1.9361, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.4534621578099839, | |
| "eval_loss": 1.939582347869873, | |
| "eval_runtime": 44.0913, | |
| "eval_samples_per_second": 5.557, | |
| "eval_steps_per_second": 2.79, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.463768115942029, | |
| "grad_norm": 3.040682077407837, | |
| "learning_rate": 5.74131823855921e-05, | |
| "loss": 2.0384, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 2.464486598968506, | |
| "learning_rate": 5.577423184847932e-05, | |
| "loss": 1.9812, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "eval_loss": 1.9403743743896484, | |
| "eval_runtime": 78.7558, | |
| "eval_samples_per_second": 3.111, | |
| "eval_steps_per_second": 1.562, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.48438003220611914, | |
| "grad_norm": 2.2205276489257812, | |
| "learning_rate": 5.4128967273616625e-05, | |
| "loss": 1.9799, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.49468599033816424, | |
| "grad_norm": 2.5809481143951416, | |
| "learning_rate": 5.247918773366112e-05, | |
| "loss": 1.9947, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.49468599033816424, | |
| "eval_loss": 1.9381928443908691, | |
| "eval_runtime": 43.5951, | |
| "eval_samples_per_second": 5.62, | |
| "eval_steps_per_second": 2.821, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5049919484702093, | |
| "grad_norm": 2.8167452812194824, | |
| "learning_rate": 5.0826697238317935e-05, | |
| "loss": 1.9157, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5152979066022544, | |
| "grad_norm": 2.8858871459960938, | |
| "learning_rate": 4.917330276168208e-05, | |
| "loss": 1.9343, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5152979066022544, | |
| "eval_loss": 1.935346007347107, | |
| "eval_runtime": 44.0224, | |
| "eval_samples_per_second": 5.565, | |
| "eval_steps_per_second": 2.794, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5256038647342995, | |
| "grad_norm": 2.6140081882476807, | |
| "learning_rate": 4.7520812266338885e-05, | |
| "loss": 2.0486, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5359098228663446, | |
| "grad_norm": 2.546766757965088, | |
| "learning_rate": 4.5871032726383386e-05, | |
| "loss": 1.9707, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5359098228663446, | |
| "eval_loss": 1.9356623888015747, | |
| "eval_runtime": 44.3481, | |
| "eval_samples_per_second": 5.524, | |
| "eval_steps_per_second": 2.774, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5462157809983897, | |
| "grad_norm": 2.5402872562408447, | |
| "learning_rate": 4.4225768151520694e-05, | |
| "loss": 1.9849, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "grad_norm": 2.748969078063965, | |
| "learning_rate": 4.2586817614407895e-05, | |
| "loss": 2.0131, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "eval_loss": 1.935119867324829, | |
| "eval_runtime": 69.0378, | |
| "eval_samples_per_second": 3.549, | |
| "eval_steps_per_second": 1.782, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5668276972624798, | |
| "grad_norm": 2.5293121337890625, | |
| "learning_rate": 4.095597328339452e-05, | |
| "loss": 2.0378, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.577133655394525, | |
| "grad_norm": 2.650944232940674, | |
| "learning_rate": 3.933501846281267e-05, | |
| "loss": 1.9416, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.577133655394525, | |
| "eval_loss": 1.9310208559036255, | |
| "eval_runtime": 265.9368, | |
| "eval_samples_per_second": 0.921, | |
| "eval_steps_per_second": 0.463, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.58743961352657, | |
| "grad_norm": 2.5562586784362793, | |
| "learning_rate": 3.772572564296005e-05, | |
| "loss": 2.0221, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5977455716586151, | |
| "grad_norm": 2.5279388427734375, | |
| "learning_rate": 3.612985456190778e-05, | |
| "loss": 1.9652, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5977455716586151, | |
| "eval_loss": 1.9351328611373901, | |
| "eval_runtime": 142.4015, | |
| "eval_samples_per_second": 1.72, | |
| "eval_steps_per_second": 0.864, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6080515297906602, | |
| "grad_norm": 2.3491861820220947, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 1.9731, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.6183574879227053, | |
| "grad_norm": 2.6937315464019775, | |
| "learning_rate": 3.298534127791785e-05, | |
| "loss": 1.9156, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6183574879227053, | |
| "eval_loss": 1.9266250133514404, | |
| "eval_runtime": 138.6395, | |
| "eval_samples_per_second": 1.767, | |
| "eval_steps_per_second": 0.887, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6286634460547504, | |
| "grad_norm": 2.3568801879882812, | |
| "learning_rate": 3.144013755408895e-05, | |
| "loss": 1.9568, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6389694041867955, | |
| "grad_norm": 2.565239906311035, | |
| "learning_rate": 2.991522876735154e-05, | |
| "loss": 1.9405, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6389694041867955, | |
| "eval_loss": 1.9259775876998901, | |
| "eval_runtime": 134.4727, | |
| "eval_samples_per_second": 1.822, | |
| "eval_steps_per_second": 0.915, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6492753623188405, | |
| "grad_norm": 2.49657940864563, | |
| "learning_rate": 2.8412282383075363e-05, | |
| "loss": 1.9631, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6595813204508857, | |
| "grad_norm": 2.4308993816375732, | |
| "learning_rate": 2.693294185106562e-05, | |
| "loss": 1.9909, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6595813204508857, | |
| "eval_loss": 1.9249815940856934, | |
| "eval_runtime": 140.3469, | |
| "eval_samples_per_second": 1.746, | |
| "eval_steps_per_second": 0.876, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6698872785829307, | |
| "grad_norm": 2.4003243446350098, | |
| "learning_rate": 2.547882480847461e-05, | |
| "loss": 1.9771, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6801932367149759, | |
| "grad_norm": 2.451406955718994, | |
| "learning_rate": 2.405152131093926e-05, | |
| "loss": 1.9179, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6801932367149759, | |
| "eval_loss": 1.9231725931167603, | |
| "eval_runtime": 263.4422, | |
| "eval_samples_per_second": 0.93, | |
| "eval_steps_per_second": 0.467, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6904991948470209, | |
| "grad_norm": 2.5658252239227295, | |
| "learning_rate": 2.2652592093878666e-05, | |
| "loss": 2.0138, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.700805152979066, | |
| "grad_norm": 2.2721314430236816, | |
| "learning_rate": 2.128356686585282e-05, | |
| "loss": 1.9877, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.700805152979066, | |
| "eval_loss": 1.921659231185913, | |
| "eval_runtime": 137.9314, | |
| "eval_samples_per_second": 1.776, | |
| "eval_steps_per_second": 0.892, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 2.5887961387634277, | |
| "learning_rate": 1.9945942635848748e-05, | |
| "loss": 1.9245, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7214170692431562, | |
| "grad_norm": 2.9357643127441406, | |
| "learning_rate": 1.8641182076323148e-05, | |
| "loss": 1.8745, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7214170692431562, | |
| "eval_loss": 1.9206682443618774, | |
| "eval_runtime": 138.8928, | |
| "eval_samples_per_second": 1.764, | |
| "eval_steps_per_second": 0.886, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7317230273752013, | |
| "grad_norm": 2.6738474369049072, | |
| "learning_rate": 1.7370711923791567e-05, | |
| "loss": 1.9979, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7420289855072464, | |
| "grad_norm": 2.5297255516052246, | |
| "learning_rate": 1.6135921418712956e-05, | |
| "loss": 2.016, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7420289855072464, | |
| "eval_loss": 1.919450044631958, | |
| "eval_runtime": 139.2841, | |
| "eval_samples_per_second": 1.759, | |
| "eval_steps_per_second": 0.883, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7523349436392914, | |
| "grad_norm": 2.4732882976531982, | |
| "learning_rate": 1.4938160786375572e-05, | |
| "loss": 1.9837, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7626409017713366, | |
| "grad_norm": 2.3005917072296143, | |
| "learning_rate": 1.3778739760445552e-05, | |
| "loss": 1.9238, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7626409017713366, | |
| "eval_loss": 1.918469786643982, | |
| "eval_runtime": 131.8213, | |
| "eval_samples_per_second": 1.859, | |
| "eval_steps_per_second": 0.933, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7729468599033816, | |
| "grad_norm": 2.417323350906372, | |
| "learning_rate": 1.2658926150792322e-05, | |
| "loss": 1.9988, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7832528180354267, | |
| "grad_norm": 2.688710927963257, | |
| "learning_rate": 1.157994445715706e-05, | |
| "loss": 1.9414, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7832528180354267, | |
| "eval_loss": 1.9193025827407837, | |
| "eval_runtime": 257.4069, | |
| "eval_samples_per_second": 0.952, | |
| "eval_steps_per_second": 0.478, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7935587761674718, | |
| "grad_norm": 2.402606248855591, | |
| "learning_rate": 1.0542974530180327e-05, | |
| "loss": 1.9017, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.8038647342995169, | |
| "grad_norm": 2.4248220920562744, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.9417, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8038647342995169, | |
| "eval_loss": 1.9172296524047852, | |
| "eval_runtime": 135.5526, | |
| "eval_samples_per_second": 1.807, | |
| "eval_steps_per_second": 0.907, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.814170692431562, | |
| "grad_norm": 2.163961172103882, | |
| "learning_rate": 8.599558442598998e-06, | |
| "loss": 1.9761, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8244766505636071, | |
| "grad_norm": 2.284884452819824, | |
| "learning_rate": 7.695237378953223e-06, | |
| "loss": 1.9647, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8244766505636071, | |
| "eval_loss": 1.916910171508789, | |
| "eval_runtime": 140.6132, | |
| "eval_samples_per_second": 1.742, | |
| "eval_steps_per_second": 0.875, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8347826086956521, | |
| "grad_norm": 2.0887887477874756, | |
| "learning_rate": 6.837175952121306e-06, | |
| "loss": 2.0037, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8450885668276973, | |
| "grad_norm": 2.22497820854187, | |
| "learning_rate": 6.026312439675552e-06, | |
| "loss": 1.9704, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8450885668276973, | |
| "eval_loss": 1.91719388961792, | |
| "eval_runtime": 138.1524, | |
| "eval_samples_per_second": 1.773, | |
| "eval_steps_per_second": 0.89, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8553945249597423, | |
| "grad_norm": 2.5339810848236084, | |
| "learning_rate": 5.263533508961827e-06, | |
| "loss": 1.9185, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8657004830917875, | |
| "grad_norm": 2.198296308517456, | |
| "learning_rate": 4.549673247541875e-06, | |
| "loss": 1.9629, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8657004830917875, | |
| "eval_loss": 1.9157441854476929, | |
| "eval_runtime": 250.478, | |
| "eval_samples_per_second": 0.978, | |
| "eval_steps_per_second": 0.491, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8760064412238325, | |
| "grad_norm": 2.4720425605773926, | |
| "learning_rate": 3.885512251130763e-06, | |
| "loss": 1.9554, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8863123993558776, | |
| "grad_norm": 2.5910909175872803, | |
| "learning_rate": 3.271776770026963e-06, | |
| "loss": 1.9574, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8863123993558776, | |
| "eval_loss": 1.9149816036224365, | |
| "eval_runtime": 128.8874, | |
| "eval_samples_per_second": 1.901, | |
| "eval_steps_per_second": 0.954, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8966183574879227, | |
| "grad_norm": 2.317239284515381, | |
| "learning_rate": 2.7091379149682685e-06, | |
| "loss": 1.895, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9069243156199678, | |
| "grad_norm": 2.7753469944000244, | |
| "learning_rate": 2.1982109232821178e-06, | |
| "loss": 1.9278, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9069243156199678, | |
| "eval_loss": 1.9143210649490356, | |
| "eval_runtime": 135.8994, | |
| "eval_samples_per_second": 1.803, | |
| "eval_steps_per_second": 0.905, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9172302737520129, | |
| "grad_norm": 2.6023426055908203, | |
| "learning_rate": 1.7395544861325718e-06, | |
| "loss": 2.0422, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.927536231884058, | |
| "grad_norm": 2.2456705570220947, | |
| "learning_rate": 1.333670137599713e-06, | |
| "loss": 2.0079, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.927536231884058, | |
| "eval_loss": 1.9140007495880127, | |
| "eval_runtime": 139.4822, | |
| "eval_samples_per_second": 1.756, | |
| "eval_steps_per_second": 0.882, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.937842190016103, | |
| "grad_norm": 2.511979103088379, | |
| "learning_rate": 9.810017062595322e-07, | |
| "loss": 1.9815, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 2.5906260013580322, | |
| "learning_rate": 6.819348298638839e-07, | |
| "loss": 1.9203, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "eval_loss": 1.9137802124023438, | |
| "eval_runtime": 137.4266, | |
| "eval_samples_per_second": 1.783, | |
| "eval_steps_per_second": 0.895, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9584541062801932, | |
| "grad_norm": 2.399129867553711, | |
| "learning_rate": 4.367965336512403e-07, | |
| "loss": 1.8967, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9687600644122383, | |
| "grad_norm": 2.576765537261963, | |
| "learning_rate": 2.458548727494292e-07, | |
| "loss": 1.9834, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9687600644122383, | |
| "eval_loss": 1.9139444828033447, | |
| "eval_runtime": 253.9256, | |
| "eval_samples_per_second": 0.965, | |
| "eval_steps_per_second": 0.484, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9790660225442834, | |
| "grad_norm": 2.6049747467041016, | |
| "learning_rate": 1.0931863906127327e-07, | |
| "loss": 1.8785, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9893719806763285, | |
| "grad_norm": 2.822763204574585, | |
| "learning_rate": 2.7337132953697554e-08, | |
| "loss": 1.8809, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9893719806763285, | |
| "eval_loss": 1.9143255949020386, | |
| "eval_runtime": 136.0427, | |
| "eval_samples_per_second": 1.801, | |
| "eval_steps_per_second": 0.904, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9996779388083736, | |
| "grad_norm": 2.4256649017333984, | |
| "learning_rate": 0.0, | |
| "loss": 1.9735, | |
| "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 | |
| } | |