| { | |
| "best_metric": 2.02008056640625, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Qwen2-7B_pct_default/checkpoint-8", | |
| "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": 5.605008125305176, | |
| "learning_rate": 3.75e-05, | |
| "loss": 2.0287, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010305958132045089, | |
| "grad_norm": 3.37660551071167, | |
| "learning_rate": 0.00015, | |
| "loss": 2.1017, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020611916264090178, | |
| "grad_norm": 3.2211694717407227, | |
| "learning_rate": 0.0003, | |
| "loss": 2.1169, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020611916264090178, | |
| "eval_loss": 2.02008056640625, | |
| "eval_runtime": 10.2397, | |
| "eval_samples_per_second": 23.927, | |
| "eval_steps_per_second": 3.027, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.030917874396135265, | |
| "grad_norm": 4.083240509033203, | |
| "learning_rate": 0.0002999179886011389, | |
| "loss": 2.0548, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.041223832528180356, | |
| "grad_norm": 4.696908473968506, | |
| "learning_rate": 0.00029967204408281613, | |
| "loss": 2.1264, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.041223832528180356, | |
| "eval_loss": 2.0886311531066895, | |
| "eval_runtime": 10.2373, | |
| "eval_samples_per_second": 23.932, | |
| "eval_steps_per_second": 3.028, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.05152979066022544, | |
| "grad_norm": 6.009027481079102, | |
| "learning_rate": 0.0002992624353817517, | |
| "loss": 2.0976, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06183574879227053, | |
| "grad_norm": 4.280951976776123, | |
| "learning_rate": 0.00029868961039904624, | |
| "loss": 2.1468, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06183574879227053, | |
| "eval_loss": 2.074465036392212, | |
| "eval_runtime": 10.269, | |
| "eval_samples_per_second": 23.858, | |
| "eval_steps_per_second": 3.019, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07214170692431562, | |
| "grad_norm": 4.368206024169922, | |
| "learning_rate": 0.00029795419551040833, | |
| "loss": 2.1692, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.08244766505636071, | |
| "grad_norm": 4.116436958312988, | |
| "learning_rate": 0.0002970569948812214, | |
| "loss": 2.0954, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08244766505636071, | |
| "eval_loss": 2.072464942932129, | |
| "eval_runtime": 10.2251, | |
| "eval_samples_per_second": 23.961, | |
| "eval_steps_per_second": 3.032, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0927536231884058, | |
| "grad_norm": 4.114132881164551, | |
| "learning_rate": 0.0002959989895872009, | |
| "loss": 2.1358, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10305958132045089, | |
| "grad_norm": 3.317972183227539, | |
| "learning_rate": 0.0002947813365416023, | |
| "loss": 2.136, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10305958132045089, | |
| "eval_loss": 2.0759308338165283, | |
| "eval_runtime": 10.1843, | |
| "eval_samples_per_second": 24.057, | |
| "eval_steps_per_second": 3.044, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11336553945249597, | |
| "grad_norm": 3.8599319458007812, | |
| "learning_rate": 0.0002934053672301536, | |
| "loss": 2.0745, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12367149758454106, | |
| "grad_norm": 14.128602027893066, | |
| "learning_rate": 0.00029187258625509513, | |
| "loss": 2.1289, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12367149758454106, | |
| "eval_loss": 2.1111183166503906, | |
| "eval_runtime": 10.1965, | |
| "eval_samples_per_second": 24.028, | |
| "eval_steps_per_second": 3.04, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13397745571658615, | |
| "grad_norm": 3.305791139602661, | |
| "learning_rate": 0.0002901846696899191, | |
| "loss": 2.1029, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14428341384863125, | |
| "grad_norm": 3.663986921310425, | |
| "learning_rate": 0.0002883434632466077, | |
| "loss": 2.131, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14428341384863125, | |
| "eval_loss": 2.07441782951355, | |
| "eval_runtime": 10.1465, | |
| "eval_samples_per_second": 24.146, | |
| "eval_steps_per_second": 3.055, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.15458937198067632, | |
| "grad_norm": 3.1207165718078613, | |
| "learning_rate": 0.00028635098025737434, | |
| "loss": 2.1223, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.16489533011272142, | |
| "grad_norm": 2.8845832347869873, | |
| "learning_rate": 0.0002842093994731145, | |
| "loss": 2.1168, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.16489533011272142, | |
| "eval_loss": 2.076624870300293, | |
| "eval_runtime": 10.2094, | |
| "eval_samples_per_second": 23.997, | |
| "eval_steps_per_second": 3.036, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1752012882447665, | |
| "grad_norm": 3.5518147945404053, | |
| "learning_rate": 0.00028192106268097334, | |
| "loss": 2.1806, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.1855072463768116, | |
| "grad_norm": 4.108978748321533, | |
| "learning_rate": 0.0002794884721436361, | |
| "loss": 2.149, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1855072463768116, | |
| "eval_loss": 2.102821111679077, | |
| "eval_runtime": 10.1389, | |
| "eval_samples_per_second": 24.164, | |
| "eval_steps_per_second": 3.058, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19581320450885667, | |
| "grad_norm": 4.722578048706055, | |
| "learning_rate": 0.0002769142878631403, | |
| "loss": 2.1607, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.20611916264090177, | |
| "grad_norm": 3.503679037094116, | |
| "learning_rate": 0.000274201324672203, | |
| "loss": 2.1947, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.20611916264090177, | |
| "eval_loss": 2.0999114513397217, | |
| "eval_runtime": 10.125, | |
| "eval_samples_per_second": 24.197, | |
| "eval_steps_per_second": 3.062, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21642512077294687, | |
| "grad_norm": 14.818785667419434, | |
| "learning_rate": 0.0002713525491562421, | |
| "loss": 2.2451, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22673107890499195, | |
| "grad_norm": 3.2971980571746826, | |
| "learning_rate": 0.00026837107640945905, | |
| "loss": 2.1727, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22673107890499195, | |
| "eval_loss": 2.099900007247925, | |
| "eval_runtime": 10.0913, | |
| "eval_samples_per_second": 24.278, | |
| "eval_steps_per_second": 3.072, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23703703703703705, | |
| "grad_norm": 3.2176926136016846, | |
| "learning_rate": 0.00026526016662852886, | |
| "loss": 2.1158, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.24734299516908212, | |
| "grad_norm": 4.07711124420166, | |
| "learning_rate": 0.0002620232215476231, | |
| "loss": 2.1438, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.24734299516908212, | |
| "eval_loss": 2.097912073135376, | |
| "eval_runtime": 10.0912, | |
| "eval_samples_per_second": 24.279, | |
| "eval_steps_per_second": 3.072, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2576489533011272, | |
| "grad_norm": 2.714487314224243, | |
| "learning_rate": 0.00025866378071866334, | |
| "loss": 2.19, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2679549114331723, | |
| "grad_norm": 3.775660276412964, | |
| "learning_rate": 0.00025518551764087326, | |
| "loss": 2.1639, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2679549114331723, | |
| "eval_loss": 2.0984129905700684, | |
| "eval_runtime": 10.0362, | |
| "eval_samples_per_second": 24.412, | |
| "eval_steps_per_second": 3.089, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2782608695652174, | |
| "grad_norm": 3.7832205295562744, | |
| "learning_rate": 0.00025159223574386114, | |
| "loss": 2.1516, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.2885668276972625, | |
| "grad_norm": 3.784287929534912, | |
| "learning_rate": 0.00024788786422862526, | |
| "loss": 2.0768, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.2885668276972625, | |
| "eval_loss": 2.096742630004883, | |
| "eval_runtime": 89.7867, | |
| "eval_samples_per_second": 2.729, | |
| "eval_steps_per_second": 0.345, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.29887278582930754, | |
| "grad_norm": 2.5659940242767334, | |
| "learning_rate": 0.00024407645377103054, | |
| "loss": 2.1725, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.30917874396135264, | |
| "grad_norm": 2.8653833866119385, | |
| "learning_rate": 0.00024016217209245374, | |
| "loss": 2.1262, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.30917874396135264, | |
| "eval_loss": 2.0943338871002197, | |
| "eval_runtime": 86.1943, | |
| "eval_samples_per_second": 2.842, | |
| "eval_steps_per_second": 0.36, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.31948470209339774, | |
| "grad_norm": 2.8778867721557617, | |
| "learning_rate": 0.0002361492994024415, | |
| "loss": 2.1553, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.32979066022544284, | |
| "grad_norm": 3.133491277694702, | |
| "learning_rate": 0.00023204222371836405, | |
| "loss": 2.1261, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.32979066022544284, | |
| "eval_loss": 2.099522113800049, | |
| "eval_runtime": 84.4047, | |
| "eval_samples_per_second": 2.903, | |
| "eval_steps_per_second": 0.367, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.34009661835748795, | |
| "grad_norm": 3.3253138065338135, | |
| "learning_rate": 0.00022784543606718227, | |
| "loss": 2.1485, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.350402576489533, | |
| "grad_norm": 3.879983901977539, | |
| "learning_rate": 0.0002235635255745762, | |
| "loss": 2.1411, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.350402576489533, | |
| "eval_loss": 2.102804660797119, | |
| "eval_runtime": 84.9029, | |
| "eval_samples_per_second": 2.886, | |
| "eval_steps_per_second": 0.365, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3607085346215781, | |
| "grad_norm": 2.706500768661499, | |
| "learning_rate": 0.00021920117444680317, | |
| "loss": 2.1075, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.3710144927536232, | |
| "grad_norm": 4.237916946411133, | |
| "learning_rate": 0.0002147631528507739, | |
| "loss": 2.1369, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.3710144927536232, | |
| "eval_loss": 2.1029863357543945, | |
| "eval_runtime": 89.0082, | |
| "eval_samples_per_second": 2.753, | |
| "eval_steps_per_second": 0.348, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.3813204508856683, | |
| "grad_norm": 2.939685344696045, | |
| "learning_rate": 0.0002102543136979454, | |
| "loss": 2.144, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.39162640901771334, | |
| "grad_norm": 2.547884941101074, | |
| "learning_rate": 0.0002056795873377331, | |
| "loss": 2.1419, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.39162640901771334, | |
| "eval_loss": 2.098917007446289, | |
| "eval_runtime": 84.8587, | |
| "eval_samples_per_second": 2.887, | |
| "eval_steps_per_second": 0.365, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.40193236714975844, | |
| "grad_norm": 3.1116697788238525, | |
| "learning_rate": 0.00020104397616624645, | |
| "loss": 2.1543, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.41223832528180354, | |
| "grad_norm": 2.5041441917419434, | |
| "learning_rate": 0.0001963525491562421, | |
| "loss": 2.165, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.41223832528180354, | |
| "eval_loss": 2.0971949100494385, | |
| "eval_runtime": 83.1216, | |
| "eval_samples_per_second": 2.947, | |
| "eval_steps_per_second": 0.373, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.42254428341384864, | |
| "grad_norm": 2.993987798690796, | |
| "learning_rate": 0.00019161043631427666, | |
| "loss": 2.1837, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43285024154589374, | |
| "grad_norm": 2.5344765186309814, | |
| "learning_rate": 0.00018682282307111987, | |
| "loss": 2.1948, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43285024154589374, | |
| "eval_loss": 2.0925116539001465, | |
| "eval_runtime": 10.2576, | |
| "eval_samples_per_second": 23.885, | |
| "eval_steps_per_second": 3.022, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.4431561996779388, | |
| "grad_norm": 2.8610734939575195, | |
| "learning_rate": 0.00018199494461156203, | |
| "loss": 2.1322, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.4534621578099839, | |
| "grad_norm": 3.8807930946350098, | |
| "learning_rate": 0.00017713208014981648, | |
| "loss": 2.1076, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.4534621578099839, | |
| "eval_loss": 2.0968382358551025, | |
| "eval_runtime": 10.2298, | |
| "eval_samples_per_second": 23.95, | |
| "eval_steps_per_second": 3.03, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.463768115942029, | |
| "grad_norm": 4.0053606033325195, | |
| "learning_rate": 0.00017223954715677627, | |
| "loss": 2.1992, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "grad_norm": 3.4194562435150146, | |
| "learning_rate": 0.00016732269554543794, | |
| "loss": 2.1183, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4740740740740741, | |
| "eval_loss": 2.0915744304656982, | |
| "eval_runtime": 10.2256, | |
| "eval_samples_per_second": 23.959, | |
| "eval_steps_per_second": 3.032, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.48438003220611914, | |
| "grad_norm": 2.9371349811553955, | |
| "learning_rate": 0.00016238690182084986, | |
| "loss": 2.1272, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.49468599033816424, | |
| "grad_norm": 3.1413886547088623, | |
| "learning_rate": 0.00015743756320098332, | |
| "loss": 2.16, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.49468599033816424, | |
| "eval_loss": 2.088491678237915, | |
| "eval_runtime": 10.2399, | |
| "eval_samples_per_second": 23.926, | |
| "eval_steps_per_second": 3.027, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5049919484702093, | |
| "grad_norm": 2.9778144359588623, | |
| "learning_rate": 0.00015248009171495378, | |
| "loss": 2.0854, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5152979066022544, | |
| "grad_norm": 2.8931424617767334, | |
| "learning_rate": 0.00014751990828504622, | |
| "loss": 2.0938, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5152979066022544, | |
| "eval_loss": 2.0884335041046143, | |
| "eval_runtime": 10.226, | |
| "eval_samples_per_second": 23.959, | |
| "eval_steps_per_second": 3.031, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5256038647342995, | |
| "grad_norm": 3.7302730083465576, | |
| "learning_rate": 0.00014256243679901663, | |
| "loss": 2.1851, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5359098228663446, | |
| "grad_norm": 2.5322744846343994, | |
| "learning_rate": 0.00013761309817915014, | |
| "loss": 2.1387, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5359098228663446, | |
| "eval_loss": 2.086639404296875, | |
| "eval_runtime": 10.1991, | |
| "eval_samples_per_second": 24.022, | |
| "eval_steps_per_second": 3.039, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5462157809983897, | |
| "grad_norm": 2.5629405975341797, | |
| "learning_rate": 0.00013267730445456208, | |
| "loss": 2.1301, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "grad_norm": 3.7714431285858154, | |
| "learning_rate": 0.00012776045284322368, | |
| "loss": 2.1735, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "eval_loss": 2.0854458808898926, | |
| "eval_runtime": 10.176, | |
| "eval_samples_per_second": 24.076, | |
| "eval_steps_per_second": 3.046, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5668276972624798, | |
| "grad_norm": 2.724277973175049, | |
| "learning_rate": 0.00012286791985018355, | |
| "loss": 2.1503, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.577133655394525, | |
| "grad_norm": 2.5124669075012207, | |
| "learning_rate": 0.00011800505538843798, | |
| "loss": 2.0786, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.577133655394525, | |
| "eval_loss": 2.0755257606506348, | |
| "eval_runtime": 10.1831, | |
| "eval_samples_per_second": 24.059, | |
| "eval_steps_per_second": 3.044, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.58743961352657, | |
| "grad_norm": 2.6235222816467285, | |
| "learning_rate": 0.00011317717692888012, | |
| "loss": 2.1536, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5977455716586151, | |
| "grad_norm": 3.0219547748565674, | |
| "learning_rate": 0.00010838956368572334, | |
| "loss": 2.0929, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5977455716586151, | |
| "eval_loss": 2.079296350479126, | |
| "eval_runtime": 10.1385, | |
| "eval_samples_per_second": 24.165, | |
| "eval_steps_per_second": 3.058, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6080515297906602, | |
| "grad_norm": 2.6528589725494385, | |
| "learning_rate": 0.0001036474508437579, | |
| "loss": 2.123, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.6183574879227053, | |
| "grad_norm": 2.549923896789551, | |
| "learning_rate": 9.895602383375353e-05, | |
| "loss": 2.0871, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6183574879227053, | |
| "eval_loss": 2.0634686946868896, | |
| "eval_runtime": 10.1669, | |
| "eval_samples_per_second": 24.098, | |
| "eval_steps_per_second": 3.049, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6286634460547504, | |
| "grad_norm": 2.451648235321045, | |
| "learning_rate": 9.432041266226686e-05, | |
| "loss": 2.0951, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6389694041867955, | |
| "grad_norm": 2.5378222465515137, | |
| "learning_rate": 8.97456863020546e-05, | |
| "loss": 2.0744, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6389694041867955, | |
| "eval_loss": 2.0636842250823975, | |
| "eval_runtime": 10.1186, | |
| "eval_samples_per_second": 24.213, | |
| "eval_steps_per_second": 3.064, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6492753623188405, | |
| "grad_norm": 2.2349236011505127, | |
| "learning_rate": 8.523684714922608e-05, | |
| "loss": 2.0909, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6595813204508857, | |
| "grad_norm": 2.813871145248413, | |
| "learning_rate": 8.079882555319684e-05, | |
| "loss": 2.1142, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6595813204508857, | |
| "eval_loss": 2.061554431915283, | |
| "eval_runtime": 10.0831, | |
| "eval_samples_per_second": 24.298, | |
| "eval_steps_per_second": 3.074, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6698872785829307, | |
| "grad_norm": 2.8577089309692383, | |
| "learning_rate": 7.643647442542382e-05, | |
| "loss": 2.1188, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6801932367149759, | |
| "grad_norm": 3.5591704845428467, | |
| "learning_rate": 7.215456393281776e-05, | |
| "loss": 2.0861, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6801932367149759, | |
| "eval_loss": 2.0569865703582764, | |
| "eval_runtime": 10.0856, | |
| "eval_samples_per_second": 24.292, | |
| "eval_steps_per_second": 3.074, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6904991948470209, | |
| "grad_norm": 2.480534791946411, | |
| "learning_rate": 6.795777628163599e-05, | |
| "loss": 2.1246, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.700805152979066, | |
| "grad_norm": 1.919311761856079, | |
| "learning_rate": 6.385070059755846e-05, | |
| "loss": 2.1428, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.700805152979066, | |
| "eval_loss": 2.0533571243286133, | |
| "eval_runtime": 85.9415, | |
| "eval_samples_per_second": 2.851, | |
| "eval_steps_per_second": 0.361, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7111111111111111, | |
| "grad_norm": 2.5054383277893066, | |
| "learning_rate": 5.983782790754623e-05, | |
| "loss": 2.0624, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7214170692431562, | |
| "grad_norm": 2.642378807067871, | |
| "learning_rate": 5.592354622896944e-05, | |
| "loss": 2.0474, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7214170692431562, | |
| "eval_loss": 2.048586130142212, | |
| "eval_runtime": 65.6601, | |
| "eval_samples_per_second": 3.731, | |
| "eval_steps_per_second": 0.472, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7317230273752013, | |
| "grad_norm": 2.2351460456848145, | |
| "learning_rate": 5.211213577137469e-05, | |
| "loss": 2.1221, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7420289855072464, | |
| "grad_norm": 2.515111207962036, | |
| "learning_rate": 4.840776425613886e-05, | |
| "loss": 2.1296, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7420289855072464, | |
| "eval_loss": 2.043865442276001, | |
| "eval_runtime": 70.6526, | |
| "eval_samples_per_second": 3.468, | |
| "eval_steps_per_second": 0.439, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7523349436392914, | |
| "grad_norm": 1.9993336200714111, | |
| "learning_rate": 4.481448235912671e-05, | |
| "loss": 2.1072, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.7626409017713366, | |
| "grad_norm": 2.0789833068847656, | |
| "learning_rate": 4.133621928133665e-05, | |
| "loss": 2.062, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7626409017713366, | |
| "eval_loss": 2.0424935817718506, | |
| "eval_runtime": 90.2963, | |
| "eval_samples_per_second": 2.713, | |
| "eval_steps_per_second": 0.343, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7729468599033816, | |
| "grad_norm": 2.315333604812622, | |
| "learning_rate": 3.797677845237696e-05, | |
| "loss": 2.1235, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7832528180354267, | |
| "grad_norm": 2.269334077835083, | |
| "learning_rate": 3.473983337147118e-05, | |
| "loss": 2.0758, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7832528180354267, | |
| "eval_loss": 2.0405173301696777, | |
| "eval_runtime": 77.3189, | |
| "eval_samples_per_second": 3.169, | |
| "eval_steps_per_second": 0.401, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7935587761674718, | |
| "grad_norm": 2.193899393081665, | |
| "learning_rate": 3.162892359054098e-05, | |
| "loss": 2.0314, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.8038647342995169, | |
| "grad_norm": 2.2756619453430176, | |
| "learning_rate": 2.8647450843757897e-05, | |
| "loss": 2.0604, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8038647342995169, | |
| "eval_loss": 2.037036180496216, | |
| "eval_runtime": 85.5713, | |
| "eval_samples_per_second": 2.863, | |
| "eval_steps_per_second": 0.362, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.814170692431562, | |
| "grad_norm": 3.295262575149536, | |
| "learning_rate": 2.5798675327796993e-05, | |
| "loss": 2.0895, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.8244766505636071, | |
| "grad_norm": 2.0809378623962402, | |
| "learning_rate": 2.3085712136859668e-05, | |
| "loss": 2.0963, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8244766505636071, | |
| "eval_loss": 2.036062717437744, | |
| "eval_runtime": 87.0694, | |
| "eval_samples_per_second": 2.814, | |
| "eval_steps_per_second": 0.356, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8347826086956521, | |
| "grad_norm": 1.9167258739471436, | |
| "learning_rate": 2.0511527856363912e-05, | |
| "loss": 2.1139, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.8450885668276973, | |
| "grad_norm": 1.832511305809021, | |
| "learning_rate": 1.8078937319026654e-05, | |
| "loss": 2.0926, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8450885668276973, | |
| "eval_loss": 2.0341949462890625, | |
| "eval_runtime": 10.2561, | |
| "eval_samples_per_second": 23.888, | |
| "eval_steps_per_second": 3.023, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8553945249597423, | |
| "grad_norm": 2.3450729846954346, | |
| "learning_rate": 1.579060052688548e-05, | |
| "loss": 2.0517, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8657004830917875, | |
| "grad_norm": 2.012261152267456, | |
| "learning_rate": 1.3649019742625623e-05, | |
| "loss": 2.0571, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8657004830917875, | |
| "eval_loss": 2.0307068824768066, | |
| "eval_runtime": 10.2233, | |
| "eval_samples_per_second": 23.965, | |
| "eval_steps_per_second": 3.032, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8760064412238325, | |
| "grad_norm": 2.20833158493042, | |
| "learning_rate": 1.1656536753392287e-05, | |
| "loss": 2.0968, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.8863123993558776, | |
| "grad_norm": 2.631930112838745, | |
| "learning_rate": 9.815330310080887e-06, | |
| "loss": 2.0858, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8863123993558776, | |
| "eval_loss": 2.0296614170074463, | |
| "eval_runtime": 10.2685, | |
| "eval_samples_per_second": 23.859, | |
| "eval_steps_per_second": 3.019, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8966183574879227, | |
| "grad_norm": 1.9802478551864624, | |
| "learning_rate": 8.127413744904804e-06, | |
| "loss": 2.0328, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9069243156199678, | |
| "grad_norm": 2.16888689994812, | |
| "learning_rate": 6.594632769846353e-06, | |
| "loss": 2.066, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9069243156199678, | |
| "eval_loss": 2.026951313018799, | |
| "eval_runtime": 10.2296, | |
| "eval_samples_per_second": 23.95, | |
| "eval_steps_per_second": 3.03, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9172302737520129, | |
| "grad_norm": 2.294292449951172, | |
| "learning_rate": 5.218663458397715e-06, | |
| "loss": 2.133, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.927536231884058, | |
| "grad_norm": 2.047563076019287, | |
| "learning_rate": 4.001010412799138e-06, | |
| "loss": 2.1284, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.927536231884058, | |
| "eval_loss": 2.026015520095825, | |
| "eval_runtime": 10.2018, | |
| "eval_samples_per_second": 24.015, | |
| "eval_steps_per_second": 3.039, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.937842190016103, | |
| "grad_norm": 2.0471081733703613, | |
| "learning_rate": 2.9430051187785962e-06, | |
| "loss": 2.0978, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "grad_norm": 2.1072356700897217, | |
| "learning_rate": 2.0458044895916513e-06, | |
| "loss": 2.0618, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9481481481481482, | |
| "eval_loss": 2.0257070064544678, | |
| "eval_runtime": 10.1953, | |
| "eval_samples_per_second": 24.031, | |
| "eval_steps_per_second": 3.041, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9584541062801932, | |
| "grad_norm": 2.2480766773223877, | |
| "learning_rate": 1.3103896009537207e-06, | |
| "loss": 2.0197, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9687600644122383, | |
| "grad_norm": 2.2486612796783447, | |
| "learning_rate": 7.375646182482875e-07, | |
| "loss": 2.1074, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9687600644122383, | |
| "eval_loss": 2.025618076324463, | |
| "eval_runtime": 10.1895, | |
| "eval_samples_per_second": 24.044, | |
| "eval_steps_per_second": 3.042, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9790660225442834, | |
| "grad_norm": 2.1844937801361084, | |
| "learning_rate": 3.2795591718381975e-07, | |
| "loss": 2.0191, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9893719806763285, | |
| "grad_norm": 2.3360085487365723, | |
| "learning_rate": 8.201139886109264e-08, | |
| "loss": 2.0625, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9893719806763285, | |
| "eval_loss": 2.025890350341797, | |
| "eval_runtime": 10.1675, | |
| "eval_samples_per_second": 24.096, | |
| "eval_steps_per_second": 3.049, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9996779388083736, | |
| "grad_norm": 2.0968825817108154, | |
| "learning_rate": 0.0, | |
| "loss": 2.1023, | |
| "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": 8.611644452078223e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |