| { | |
| "best_metric": 2.2199668884277344, | |
| "best_model_checkpoint": "/home/datta0/models/lora_final/Meta-Llama-3-8B_pct_ortho/checkpoint-328", | |
| "epoch": 0.9990344383649823, | |
| "eval_steps": 8, | |
| "global_step": 388, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002574831026713872, | |
| "grad_norm": 4.8250555992126465, | |
| "learning_rate": 3.75e-05, | |
| "loss": 2.3613, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.010299324106855488, | |
| "grad_norm": 3.812638998031616, | |
| "learning_rate": 0.00015, | |
| "loss": 2.3464, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.020598648213710977, | |
| "grad_norm": 4.135324478149414, | |
| "learning_rate": 0.0003, | |
| "loss": 2.2594, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.020598648213710977, | |
| "eval_loss": 2.2432966232299805, | |
| "eval_runtime": 10.4685, | |
| "eval_samples_per_second": 23.404, | |
| "eval_steps_per_second": 2.961, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.03089797232056646, | |
| "grad_norm": 3.940943956375122, | |
| "learning_rate": 0.0002999179886011389, | |
| "loss": 2.2353, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.04119729642742195, | |
| "grad_norm": 3.0164599418640137, | |
| "learning_rate": 0.00029967204408281613, | |
| "loss": 2.2616, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04119729642742195, | |
| "eval_loss": 2.2376890182495117, | |
| "eval_runtime": 10.472, | |
| "eval_samples_per_second": 23.396, | |
| "eval_steps_per_second": 2.96, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.051496620534277435, | |
| "grad_norm": 3.438307046890259, | |
| "learning_rate": 0.0002992624353817517, | |
| "loss": 2.2166, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.06179594464113292, | |
| "grad_norm": 3.0681569576263428, | |
| "learning_rate": 0.00029868961039904624, | |
| "loss": 2.2041, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06179594464113292, | |
| "eval_loss": 2.2342782020568848, | |
| "eval_runtime": 10.4748, | |
| "eval_samples_per_second": 23.389, | |
| "eval_steps_per_second": 2.959, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.07209526874798841, | |
| "grad_norm": 2.55240535736084, | |
| "learning_rate": 0.00029795419551040833, | |
| "loss": 2.2296, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0823945928548439, | |
| "grad_norm": 2.451361894607544, | |
| "learning_rate": 0.0002970569948812214, | |
| "loss": 2.2802, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0823945928548439, | |
| "eval_loss": 2.253261089324951, | |
| "eval_runtime": 10.4553, | |
| "eval_samples_per_second": 23.433, | |
| "eval_steps_per_second": 2.965, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09269391696169939, | |
| "grad_norm": 5.104744911193848, | |
| "learning_rate": 0.0002959989895872009, | |
| "loss": 2.2541, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10299324106855487, | |
| "grad_norm": 2.3457939624786377, | |
| "learning_rate": 0.0002947813365416023, | |
| "loss": 2.2832, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10299324106855487, | |
| "eval_loss": 2.2618255615234375, | |
| "eval_runtime": 10.4563, | |
| "eval_samples_per_second": 23.431, | |
| "eval_steps_per_second": 2.965, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.11329256517541036, | |
| "grad_norm": 3.244114398956299, | |
| "learning_rate": 0.0002934053672301536, | |
| "loss": 2.2859, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.12359188928226585, | |
| "grad_norm": 2.5838561058044434, | |
| "learning_rate": 0.00029187258625509513, | |
| "loss": 2.2484, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12359188928226585, | |
| "eval_loss": 2.25663685798645, | |
| "eval_runtime": 10.4533, | |
| "eval_samples_per_second": 23.438, | |
| "eval_steps_per_second": 2.966, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.13389121338912133, | |
| "grad_norm": 2.609206199645996, | |
| "learning_rate": 0.0002901846696899191, | |
| "loss": 2.3116, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.14419053749597682, | |
| "grad_norm": 3.638091802597046, | |
| "learning_rate": 0.0002883434632466077, | |
| "loss": 2.2735, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.14419053749597682, | |
| "eval_loss": 2.2576816082000732, | |
| "eval_runtime": 10.4026, | |
| "eval_samples_per_second": 23.552, | |
| "eval_steps_per_second": 2.98, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.15448986160283232, | |
| "grad_norm": 2.5911107063293457, | |
| "learning_rate": 0.00028635098025737434, | |
| "loss": 2.2803, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.1647891857096878, | |
| "grad_norm": 2.4840705394744873, | |
| "learning_rate": 0.0002842093994731145, | |
| "loss": 2.293, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.1647891857096878, | |
| "eval_loss": 2.2783665657043457, | |
| "eval_runtime": 10.3852, | |
| "eval_samples_per_second": 23.591, | |
| "eval_steps_per_second": 2.985, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.17508850981654328, | |
| "grad_norm": 2.24977970123291, | |
| "learning_rate": 0.00028192106268097334, | |
| "loss": 2.3169, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.18538783392339878, | |
| "grad_norm": 2.666362762451172, | |
| "learning_rate": 0.0002794884721436361, | |
| "loss": 2.2518, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.18538783392339878, | |
| "eval_loss": 2.2818338871002197, | |
| "eval_runtime": 10.3527, | |
| "eval_samples_per_second": 23.665, | |
| "eval_steps_per_second": 2.994, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.19568715803025427, | |
| "grad_norm": 2.480801582336426, | |
| "learning_rate": 0.0002769142878631403, | |
| "loss": 2.3334, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 0.20598648213710974, | |
| "grad_norm": 2.343674898147583, | |
| "learning_rate": 0.000274201324672203, | |
| "loss": 2.2922, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.20598648213710974, | |
| "eval_loss": 2.291419744491577, | |
| "eval_runtime": 10.3695, | |
| "eval_samples_per_second": 23.627, | |
| "eval_steps_per_second": 2.99, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.21628580624396523, | |
| "grad_norm": 2.1927764415740967, | |
| "learning_rate": 0.0002713525491562421, | |
| "loss": 2.2794, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.22658513035082073, | |
| "grad_norm": 2.764920711517334, | |
| "learning_rate": 0.00026837107640945905, | |
| "loss": 2.3199, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.22658513035082073, | |
| "eval_loss": 2.2720041275024414, | |
| "eval_runtime": 10.3284, | |
| "eval_samples_per_second": 23.721, | |
| "eval_steps_per_second": 3.001, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.23688445445767622, | |
| "grad_norm": 2.5399646759033203, | |
| "learning_rate": 0.00026526016662852886, | |
| "loss": 2.2966, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 0.2471837785645317, | |
| "grad_norm": 2.4544336795806885, | |
| "learning_rate": 0.0002620232215476231, | |
| "loss": 2.3807, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2471837785645317, | |
| "eval_loss": 2.278792142868042, | |
| "eval_runtime": 10.3488, | |
| "eval_samples_per_second": 23.674, | |
| "eval_steps_per_second": 2.996, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.2574831026713872, | |
| "grad_norm": 2.6318697929382324, | |
| "learning_rate": 0.00025866378071866334, | |
| "loss": 2.3591, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26778242677824265, | |
| "grad_norm": 2.789501667022705, | |
| "learning_rate": 0.00025518551764087326, | |
| "loss": 2.3528, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.26778242677824265, | |
| "eval_loss": 2.2830569744110107, | |
| "eval_runtime": 10.2996, | |
| "eval_samples_per_second": 23.787, | |
| "eval_steps_per_second": 3.01, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 0.2780817508850982, | |
| "grad_norm": 2.526961326599121, | |
| "learning_rate": 0.00025159223574386114, | |
| "loss": 2.3028, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.28838107499195365, | |
| "grad_norm": 3.0287418365478516, | |
| "learning_rate": 0.00024788786422862526, | |
| "loss": 2.3144, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.28838107499195365, | |
| "eval_loss": 2.295750379562378, | |
| "eval_runtime": 89.6533, | |
| "eval_samples_per_second": 2.733, | |
| "eval_steps_per_second": 0.346, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 0.29868039909880917, | |
| "grad_norm": 3.0724334716796875, | |
| "learning_rate": 0.00024407645377103054, | |
| "loss": 2.3251, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 0.30897972320566464, | |
| "grad_norm": 3.3411991596221924, | |
| "learning_rate": 0.00024016217209245374, | |
| "loss": 2.3652, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.30897972320566464, | |
| "eval_loss": 2.295003652572632, | |
| "eval_runtime": 92.6572, | |
| "eval_samples_per_second": 2.644, | |
| "eval_steps_per_second": 0.335, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.3192790473125201, | |
| "grad_norm": 2.637554407119751, | |
| "learning_rate": 0.0002361492994024415, | |
| "loss": 2.2994, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 0.3295783714193756, | |
| "grad_norm": 2.461324453353882, | |
| "learning_rate": 0.00023204222371836405, | |
| "loss": 2.3637, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3295783714193756, | |
| "eval_loss": 2.2845101356506348, | |
| "eval_runtime": 94.8437, | |
| "eval_samples_per_second": 2.583, | |
| "eval_steps_per_second": 0.327, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 0.3398776955262311, | |
| "grad_norm": 2.624124765396118, | |
| "learning_rate": 0.00022784543606718227, | |
| "loss": 2.332, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.35017701963308656, | |
| "grad_norm": 2.3923728466033936, | |
| "learning_rate": 0.0002235635255745762, | |
| "loss": 2.3014, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.35017701963308656, | |
| "eval_loss": 2.278064250946045, | |
| "eval_runtime": 101.7542, | |
| "eval_samples_per_second": 2.408, | |
| "eval_steps_per_second": 0.305, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 0.3604763437399421, | |
| "grad_norm": 3.1025068759918213, | |
| "learning_rate": 0.00021920117444680317, | |
| "loss": 2.3819, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.37077566784679755, | |
| "grad_norm": 2.671719551086426, | |
| "learning_rate": 0.0002147631528507739, | |
| "loss": 2.3067, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.37077566784679755, | |
| "eval_loss": 2.2829501628875732, | |
| "eval_runtime": 101.7933, | |
| "eval_samples_per_second": 2.407, | |
| "eval_steps_per_second": 0.305, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.381074991953653, | |
| "grad_norm": 2.460653781890869, | |
| "learning_rate": 0.0002102543136979454, | |
| "loss": 2.2942, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 0.39137431606050854, | |
| "grad_norm": 2.579962730407715, | |
| "learning_rate": 0.0002056795873377331, | |
| "loss": 2.3242, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.39137431606050854, | |
| "eval_loss": 2.2787716388702393, | |
| "eval_runtime": 105.0919, | |
| "eval_samples_per_second": 2.331, | |
| "eval_steps_per_second": 0.295, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 0.401673640167364, | |
| "grad_norm": 2.271052837371826, | |
| "learning_rate": 0.00020104397616624645, | |
| "loss": 2.3009, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.4119729642742195, | |
| "grad_norm": 3.09191632270813, | |
| "learning_rate": 0.0001963525491562421, | |
| "loss": 2.3184, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.4119729642742195, | |
| "eval_loss": 2.265949249267578, | |
| "eval_runtime": 100.9124, | |
| "eval_samples_per_second": 2.428, | |
| "eval_steps_per_second": 0.307, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.422272288381075, | |
| "grad_norm": 2.4084908962249756, | |
| "learning_rate": 0.00019161043631427666, | |
| "loss": 2.2862, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 0.43257161248793047, | |
| "grad_norm": 2.453981399536133, | |
| "learning_rate": 0.00018682282307111987, | |
| "loss": 2.3574, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.43257161248793047, | |
| "eval_loss": 2.278163433074951, | |
| "eval_runtime": 10.4957, | |
| "eval_samples_per_second": 23.343, | |
| "eval_steps_per_second": 2.954, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.442870936594786, | |
| "grad_norm": 2.5125510692596436, | |
| "learning_rate": 0.00018199494461156203, | |
| "loss": 2.3157, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 0.45317026070164146, | |
| "grad_norm": 3.002711057662964, | |
| "learning_rate": 0.00017713208014981648, | |
| "loss": 2.3006, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.45317026070164146, | |
| "eval_loss": 2.2733473777770996, | |
| "eval_runtime": 10.4821, | |
| "eval_samples_per_second": 23.373, | |
| "eval_steps_per_second": 2.957, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 0.4634695848084969, | |
| "grad_norm": 2.4078829288482666, | |
| "learning_rate": 0.00017223954715677627, | |
| "loss": 2.3307, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.47376890891535245, | |
| "grad_norm": 2.81722092628479, | |
| "learning_rate": 0.00016732269554543794, | |
| "loss": 2.3082, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.47376890891535245, | |
| "eval_loss": 2.2698850631713867, | |
| "eval_runtime": 10.4759, | |
| "eval_samples_per_second": 23.387, | |
| "eval_steps_per_second": 2.959, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 0.4840682330222079, | |
| "grad_norm": 2.851599931716919, | |
| "learning_rate": 0.00016238690182084986, | |
| "loss": 2.2678, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 0.4943675571290634, | |
| "grad_norm": 3.283416271209717, | |
| "learning_rate": 0.00015743756320098332, | |
| "loss": 2.3097, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.4943675571290634, | |
| "eval_loss": 2.2615249156951904, | |
| "eval_runtime": 10.4874, | |
| "eval_samples_per_second": 23.361, | |
| "eval_steps_per_second": 2.956, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.5046668812359189, | |
| "grad_norm": 2.369040012359619, | |
| "learning_rate": 0.00015248009171495378, | |
| "loss": 2.304, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 0.5149662053427744, | |
| "grad_norm": 2.1018686294555664, | |
| "learning_rate": 0.00014751990828504622, | |
| "loss": 2.3003, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5149662053427744, | |
| "eval_loss": 2.264901638031006, | |
| "eval_runtime": 10.4653, | |
| "eval_samples_per_second": 23.411, | |
| "eval_steps_per_second": 2.962, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5252655294496299, | |
| "grad_norm": 2.4146642684936523, | |
| "learning_rate": 0.00014256243679901663, | |
| "loss": 2.2735, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.5355648535564853, | |
| "grad_norm": 2.4833011627197266, | |
| "learning_rate": 0.00013761309817915014, | |
| "loss": 2.3027, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5355648535564853, | |
| "eval_loss": 2.2593843936920166, | |
| "eval_runtime": 10.4677, | |
| "eval_samples_per_second": 23.405, | |
| "eval_steps_per_second": 2.961, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 0.5458641776633408, | |
| "grad_norm": 2.3660264015197754, | |
| "learning_rate": 0.00013267730445456208, | |
| "loss": 2.2475, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 0.5561635017701964, | |
| "grad_norm": 2.8252508640289307, | |
| "learning_rate": 0.00012776045284322368, | |
| "loss": 2.3262, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5561635017701964, | |
| "eval_loss": 2.249122142791748, | |
| "eval_runtime": 10.4377, | |
| "eval_samples_per_second": 23.473, | |
| "eval_steps_per_second": 2.97, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.5664628258770518, | |
| "grad_norm": 2.3867785930633545, | |
| "learning_rate": 0.00012286791985018355, | |
| "loss": 2.2732, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.5767621499839073, | |
| "grad_norm": 2.2431817054748535, | |
| "learning_rate": 0.00011800505538843798, | |
| "loss": 2.3118, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5767621499839073, | |
| "eval_loss": 2.2598836421966553, | |
| "eval_runtime": 10.3894, | |
| "eval_samples_per_second": 23.582, | |
| "eval_steps_per_second": 2.984, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 0.5870614740907628, | |
| "grad_norm": 2.2373931407928467, | |
| "learning_rate": 0.00011317717692888012, | |
| "loss": 2.3041, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 0.5973607981976183, | |
| "grad_norm": 2.278763771057129, | |
| "learning_rate": 0.00010838956368572334, | |
| "loss": 2.2904, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.5973607981976183, | |
| "eval_loss": 2.26226544380188, | |
| "eval_runtime": 10.39, | |
| "eval_samples_per_second": 23.58, | |
| "eval_steps_per_second": 2.984, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 0.6076601223044737, | |
| "grad_norm": 2.2224009037017822, | |
| "learning_rate": 0.0001036474508437579, | |
| "loss": 2.3356, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 0.6179594464113293, | |
| "grad_norm": 1.9502872228622437, | |
| "learning_rate": 9.895602383375353e-05, | |
| "loss": 2.2519, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6179594464113293, | |
| "eval_loss": 2.249502658843994, | |
| "eval_runtime": 10.3323, | |
| "eval_samples_per_second": 23.712, | |
| "eval_steps_per_second": 3.0, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.6282587705181848, | |
| "grad_norm": 1.9756816625595093, | |
| "learning_rate": 9.432041266226686e-05, | |
| "loss": 2.3011, | |
| "step": 244 | |
| }, | |
| { | |
| "epoch": 0.6385580946250402, | |
| "grad_norm": 2.24442195892334, | |
| "learning_rate": 8.97456863020546e-05, | |
| "loss": 2.2907, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6385580946250402, | |
| "eval_loss": 2.252591848373413, | |
| "eval_runtime": 10.375, | |
| "eval_samples_per_second": 23.615, | |
| "eval_steps_per_second": 2.988, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 0.6488574187318957, | |
| "grad_norm": 2.2955076694488525, | |
| "learning_rate": 8.523684714922608e-05, | |
| "loss": 2.2493, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 0.6591567428387513, | |
| "grad_norm": 2.3158631324768066, | |
| "learning_rate": 8.079882555319684e-05, | |
| "loss": 2.2864, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6591567428387513, | |
| "eval_loss": 2.25117826461792, | |
| "eval_runtime": 10.3184, | |
| "eval_samples_per_second": 23.744, | |
| "eval_steps_per_second": 3.004, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 0.6694560669456067, | |
| "grad_norm": 1.8003387451171875, | |
| "learning_rate": 7.643647442542382e-05, | |
| "loss": 2.2806, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.6797553910524622, | |
| "grad_norm": 2.0942351818084717, | |
| "learning_rate": 7.215456393281776e-05, | |
| "loss": 2.242, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6797553910524622, | |
| "eval_loss": 2.249201774597168, | |
| "eval_runtime": 10.2986, | |
| "eval_samples_per_second": 23.79, | |
| "eval_steps_per_second": 3.01, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 0.6900547151593177, | |
| "grad_norm": 2.113708019256592, | |
| "learning_rate": 6.795777628163599e-05, | |
| "loss": 2.2817, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.7003540392661731, | |
| "grad_norm": 2.8915371894836426, | |
| "learning_rate": 6.385070059755846e-05, | |
| "loss": 2.2941, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7003540392661731, | |
| "eval_loss": 2.2414655685424805, | |
| "eval_runtime": 81.0572, | |
| "eval_samples_per_second": 3.023, | |
| "eval_steps_per_second": 0.382, | |
| "step": 272 | |
| }, | |
| { | |
| "epoch": 0.7106533633730286, | |
| "grad_norm": 2.2395060062408447, | |
| "learning_rate": 5.983782790754623e-05, | |
| "loss": 2.2913, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 0.7209526874798842, | |
| "grad_norm": 2.083782911300659, | |
| "learning_rate": 5.592354622896944e-05, | |
| "loss": 2.2799, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7209526874798842, | |
| "eval_loss": 2.238328695297241, | |
| "eval_runtime": 47.7655, | |
| "eval_samples_per_second": 5.129, | |
| "eval_steps_per_second": 0.649, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.7312520115867396, | |
| "grad_norm": 2.176422595977783, | |
| "learning_rate": 5.211213577137469e-05, | |
| "loss": 2.2807, | |
| "step": 284 | |
| }, | |
| { | |
| "epoch": 0.7415513356935951, | |
| "grad_norm": 2.0646986961364746, | |
| "learning_rate": 4.840776425613886e-05, | |
| "loss": 2.2881, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7415513356935951, | |
| "eval_loss": 2.235785484313965, | |
| "eval_runtime": 48.1216, | |
| "eval_samples_per_second": 5.091, | |
| "eval_steps_per_second": 0.644, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 0.7518506598004506, | |
| "grad_norm": 1.7966443300247192, | |
| "learning_rate": 4.481448235912671e-05, | |
| "loss": 2.2754, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 0.762149983907306, | |
| "grad_norm": 1.8745654821395874, | |
| "learning_rate": 4.133621928133665e-05, | |
| "loss": 2.2797, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.762149983907306, | |
| "eval_loss": 2.2380549907684326, | |
| "eval_runtime": 51.7197, | |
| "eval_samples_per_second": 4.737, | |
| "eval_steps_per_second": 0.599, | |
| "step": 296 | |
| }, | |
| { | |
| "epoch": 0.7724493080141616, | |
| "grad_norm": 2.07661771774292, | |
| "learning_rate": 3.797677845237696e-05, | |
| "loss": 2.2368, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7827486321210171, | |
| "grad_norm": 2.3376801013946533, | |
| "learning_rate": 3.473983337147118e-05, | |
| "loss": 2.3197, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7827486321210171, | |
| "eval_loss": 2.2254865169525146, | |
| "eval_runtime": 50.4873, | |
| "eval_samples_per_second": 4.853, | |
| "eval_steps_per_second": 0.614, | |
| "step": 304 | |
| }, | |
| { | |
| "epoch": 0.7930479562278725, | |
| "grad_norm": 2.3710439205169678, | |
| "learning_rate": 3.162892359054098e-05, | |
| "loss": 2.2302, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 0.803347280334728, | |
| "grad_norm": 2.251873731613159, | |
| "learning_rate": 2.8647450843757897e-05, | |
| "loss": 2.2507, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.803347280334728, | |
| "eval_loss": 2.228350877761841, | |
| "eval_runtime": 84.236, | |
| "eval_samples_per_second": 2.908, | |
| "eval_steps_per_second": 0.368, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 0.8136466044415835, | |
| "grad_norm": 2.0786876678466797, | |
| "learning_rate": 2.5798675327796993e-05, | |
| "loss": 2.2143, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 0.823945928548439, | |
| "grad_norm": 2.0034406185150146, | |
| "learning_rate": 2.3085712136859668e-05, | |
| "loss": 2.236, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.823945928548439, | |
| "eval_loss": 2.2312803268432617, | |
| "eval_runtime": 81.5754, | |
| "eval_samples_per_second": 3.003, | |
| "eval_steps_per_second": 0.38, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.8342452526552945, | |
| "grad_norm": 1.921857476234436, | |
| "learning_rate": 2.0511527856363912e-05, | |
| "loss": 2.2961, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.84454457676215, | |
| "grad_norm": 1.6842375993728638, | |
| "learning_rate": 1.8078937319026654e-05, | |
| "loss": 2.2667, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.84454457676215, | |
| "eval_loss": 2.2199668884277344, | |
| "eval_runtime": 10.4841, | |
| "eval_samples_per_second": 23.369, | |
| "eval_steps_per_second": 2.957, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 0.8548439008690055, | |
| "grad_norm": 1.776142954826355, | |
| "learning_rate": 1.579060052688548e-05, | |
| "loss": 2.2333, | |
| "step": 332 | |
| }, | |
| { | |
| "epoch": 0.8651432249758609, | |
| "grad_norm": 2.0090060234069824, | |
| "learning_rate": 1.3649019742625623e-05, | |
| "loss": 2.2763, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8651432249758609, | |
| "eval_loss": 2.2255115509033203, | |
| "eval_runtime": 10.47, | |
| "eval_samples_per_second": 23.4, | |
| "eval_steps_per_second": 2.961, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 0.8754425490827165, | |
| "grad_norm": 2.0487890243530273, | |
| "learning_rate": 1.1656536753392287e-05, | |
| "loss": 2.2732, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.885741873189572, | |
| "grad_norm": 1.6611812114715576, | |
| "learning_rate": 9.815330310080887e-06, | |
| "loss": 2.2915, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.885741873189572, | |
| "eval_loss": 2.2228875160217285, | |
| "eval_runtime": 10.4667, | |
| "eval_samples_per_second": 23.407, | |
| "eval_steps_per_second": 2.962, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 0.8960411972964274, | |
| "grad_norm": 1.7494527101516724, | |
| "learning_rate": 8.127413744904804e-06, | |
| "loss": 2.2629, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 0.9063405214032829, | |
| "grad_norm": 2.0750112533569336, | |
| "learning_rate": 6.594632769846353e-06, | |
| "loss": 2.2554, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9063405214032829, | |
| "eval_loss": 2.221097230911255, | |
| "eval_runtime": 10.4622, | |
| "eval_samples_per_second": 23.418, | |
| "eval_steps_per_second": 2.963, | |
| "step": 352 | |
| }, | |
| { | |
| "epoch": 0.9166398455101384, | |
| "grad_norm": 1.803432583808899, | |
| "learning_rate": 5.218663458397715e-06, | |
| "loss": 2.2508, | |
| "step": 356 | |
| }, | |
| { | |
| "epoch": 0.9269391696169939, | |
| "grad_norm": 1.7050764560699463, | |
| "learning_rate": 4.001010412799138e-06, | |
| "loss": 2.2237, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9269391696169939, | |
| "eval_loss": 2.224059820175171, | |
| "eval_runtime": 10.5244, | |
| "eval_samples_per_second": 23.279, | |
| "eval_steps_per_second": 2.946, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.9372384937238494, | |
| "grad_norm": 1.768523097038269, | |
| "learning_rate": 2.9430051187785962e-06, | |
| "loss": 2.2772, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 0.9475378178307049, | |
| "grad_norm": 1.7917652130126953, | |
| "learning_rate": 2.0458044895916513e-06, | |
| "loss": 2.2446, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9475378178307049, | |
| "eval_loss": 2.2229297161102295, | |
| "eval_runtime": 10.4777, | |
| "eval_samples_per_second": 23.383, | |
| "eval_steps_per_second": 2.959, | |
| "step": 368 | |
| }, | |
| { | |
| "epoch": 0.9578371419375603, | |
| "grad_norm": 1.6179791688919067, | |
| "learning_rate": 1.3103896009537207e-06, | |
| "loss": 2.2329, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 0.9681364660444158, | |
| "grad_norm": 1.8671883344650269, | |
| "learning_rate": 7.375646182482875e-07, | |
| "loss": 2.2926, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9681364660444158, | |
| "eval_loss": 2.2224321365356445, | |
| "eval_runtime": 10.4643, | |
| "eval_samples_per_second": 23.413, | |
| "eval_steps_per_second": 2.962, | |
| "step": 376 | |
| }, | |
| { | |
| "epoch": 0.9784357901512714, | |
| "grad_norm": 2.0502476692199707, | |
| "learning_rate": 3.2795591718381975e-07, | |
| "loss": 2.2313, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.9887351142581268, | |
| "grad_norm": 1.8618476390838623, | |
| "learning_rate": 8.201139886109264e-08, | |
| "loss": 2.2813, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9887351142581268, | |
| "eval_loss": 2.2218899726867676, | |
| "eval_runtime": 10.462, | |
| "eval_samples_per_second": 23.418, | |
| "eval_steps_per_second": 2.963, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 0.9990344383649823, | |
| "grad_norm": 1.9400445222854614, | |
| "learning_rate": 0.0, | |
| "loss": 2.2718, | |
| "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": 9.087198145554678e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |