| { | |
| "best_metric": 0.6952534317970276, | |
| "best_model_checkpoint": "/home/llmadmin/models/loras-trained/wizardLM-llama-lora-13b/checkpoint-1200", | |
| "epoch": 2.1176470588235294, | |
| "global_step": 1200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 2.6999999999999996e-05, | |
| "loss": 1.2009, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 5.6999999999999996e-05, | |
| "loss": 1.198, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 8.699999999999999e-05, | |
| "loss": 1.1318, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.000117, | |
| "loss": 1.0446, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.000147, | |
| "loss": 0.9254, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 0.00017699999999999997, | |
| "loss": 0.8364, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.00020699999999999996, | |
| "loss": 0.818, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 0.000237, | |
| "loss": 0.8094, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 0.000267, | |
| "loss": 0.7891, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00029699999999999996, | |
| "loss": 0.7915, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.0002983103879849812, | |
| "loss": 0.777, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.000296433041301627, | |
| "loss": 0.7809, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0002945556946182728, | |
| "loss": 0.7705, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 0.0002926783479349186, | |
| "loss": 0.7545, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.00029080100125156443, | |
| "loss": 0.7584, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 0.00028892365456821025, | |
| "loss": 0.751, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00028704630788485607, | |
| "loss": 0.757, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 0.00028516896120150183, | |
| "loss": 0.7446, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00028329161451814765, | |
| "loss": 0.7411, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0002814142678347935, | |
| "loss": 0.754, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 0.7339245676994324, | |
| "eval_runtime": 65.2861, | |
| "eval_samples_per_second": 30.634, | |
| "eval_steps_per_second": 0.766, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 0.00027953692115143924, | |
| "loss": 0.7487, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 0.00027765957446808506, | |
| "loss": 0.7478, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 0.0002757822277847309, | |
| "loss": 0.7336, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0002739048811013767, | |
| "loss": 0.7386, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002720275344180225, | |
| "loss": 0.7359, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0002701501877346683, | |
| "loss": 0.7455, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 0.0002682728410513141, | |
| "loss": 0.733, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 0.0002663954943679599, | |
| "loss": 0.7332, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.00026451814768460575, | |
| "loss": 0.7253, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00026264080100125157, | |
| "loss": 0.732, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 0.00026076345431789733, | |
| "loss": 0.7271, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 0.00025888610763454315, | |
| "loss": 0.7365, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.00025700876095118897, | |
| "loss": 0.7266, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0002551314142678348, | |
| "loss": 0.733, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00025325406758448056, | |
| "loss": 0.7343, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 0.0002513767209011264, | |
| "loss": 0.723, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.0002494993742177722, | |
| "loss": 0.7391, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 0.000247622027534418, | |
| "loss": 0.7265, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00024574468085106384, | |
| "loss": 0.7283, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0002438673341677096, | |
| "loss": 0.7306, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 0.7154878377914429, | |
| "eval_runtime": 65.4382, | |
| "eval_samples_per_second": 30.563, | |
| "eval_steps_per_second": 0.764, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00024198998748435542, | |
| "loss": 0.725, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00024011264080100122, | |
| "loss": 0.7275, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 0.000238235294117647, | |
| "loss": 0.7188, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "learning_rate": 0.00023635794743429286, | |
| "loss": 0.7216, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 0.00023448060075093865, | |
| "loss": 0.7132, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00023260325406758447, | |
| "loss": 0.7372, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 0.00023072590738423026, | |
| "loss": 0.7111, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.00022884856070087608, | |
| "loss": 0.7131, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.00022697121401752188, | |
| "loss": 0.7193, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.00022509386733416767, | |
| "loss": 0.7152, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00022321652065081352, | |
| "loss": 0.7234, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0002213391739674593, | |
| "loss": 0.7193, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00021946182728410513, | |
| "loss": 0.7208, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.00021758448060075092, | |
| "loss": 0.7205, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "learning_rate": 0.00021570713391739672, | |
| "loss": 0.7154, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.00021382978723404254, | |
| "loss": 0.7224, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 0.00021195244055068833, | |
| "loss": 0.7015, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.00021007509386733418, | |
| "loss": 0.7168, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.00020819774718397997, | |
| "loss": 0.7281, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.00020632040050062576, | |
| "loss": 0.7112, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_loss": 0.7068994045257568, | |
| "eval_runtime": 65.2314, | |
| "eval_samples_per_second": 30.66, | |
| "eval_steps_per_second": 0.767, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00020444305381727158, | |
| "loss": 0.7133, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 0.00020256570713391738, | |
| "loss": 0.7047, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 0.00020068836045056317, | |
| "loss": 0.7163, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 0.000198811013767209, | |
| "loss": 0.7168, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.00019693366708385478, | |
| "loss": 0.7028, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 0.00019505632040050063, | |
| "loss": 0.7064, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00019317897371714642, | |
| "loss": 0.7127, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.00019130162703379222, | |
| "loss": 0.7193, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 0.00018942428035043804, | |
| "loss": 0.6966, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00018754693366708383, | |
| "loss": 0.7104, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.00018566958698372962, | |
| "loss": 0.7076, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.00018379224030037544, | |
| "loss": 0.6971, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00018191489361702126, | |
| "loss": 0.6941, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.00018003754693366708, | |
| "loss": 0.6992, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "learning_rate": 0.00017816020025031287, | |
| "loss": 0.7059, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 0.0001762828535669587, | |
| "loss": 0.6936, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "learning_rate": 0.0001744055068836045, | |
| "loss": 0.6961, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00017252816020025028, | |
| "loss": 0.7125, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 0.0001706508135168961, | |
| "loss": 0.7046, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 0.00016877346683354192, | |
| "loss": 0.7034, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_loss": 0.7016428709030151, | |
| "eval_runtime": 65.2245, | |
| "eval_samples_per_second": 30.663, | |
| "eval_steps_per_second": 0.767, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "learning_rate": 0.00016689612015018774, | |
| "loss": 0.7011, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.00016501877346683353, | |
| "loss": 0.7049, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.00016314142678347933, | |
| "loss": 0.6965, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.00016126408010012515, | |
| "loss": 0.6965, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.00015938673341677094, | |
| "loss": 0.6975, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 0.00015750938673341673, | |
| "loss": 0.7094, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00015563204005006258, | |
| "loss": 0.6918, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.00015375469336670837, | |
| "loss": 0.7031, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "learning_rate": 0.0001518773466833542, | |
| "loss": 0.7027, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 0.00015, | |
| "loss": 0.6996, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 0.00014812265331664578, | |
| "loss": 0.7048, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.0001462453066332916, | |
| "loss": 0.6962, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.00014436795994993742, | |
| "loss": 0.6918, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 0.0001424906132665832, | |
| "loss": 0.6979, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 0.000140613266583229, | |
| "loss": 0.6958, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00013873591989987483, | |
| "loss": 0.7027, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.00013685857321652065, | |
| "loss": 0.7092, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.00013498122653316644, | |
| "loss": 0.6917, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00013310387984981226, | |
| "loss": 0.6996, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "learning_rate": 0.00013122653316645805, | |
| "loss": 0.6889, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "eval_loss": 0.6975076794624329, | |
| "eval_runtime": 65.2345, | |
| "eval_samples_per_second": 30.659, | |
| "eval_steps_per_second": 0.766, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.00012934918648310387, | |
| "loss": 0.7083, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.00012747183979974967, | |
| "loss": 0.6934, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 0.00012559449311639549, | |
| "loss": 0.7002, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 0.00012371714643304128, | |
| "loss": 0.6899, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0001218397997496871, | |
| "loss": 0.6856, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.00011996245306633289, | |
| "loss": 0.695, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00011808510638297871, | |
| "loss": 0.6934, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "learning_rate": 0.00011620775969962452, | |
| "loss": 0.6933, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.00011433041301627033, | |
| "loss": 0.6983, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 0.00011245306633291615, | |
| "loss": 0.705, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.00011057571964956194, | |
| "loss": 0.6944, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.00010869837296620774, | |
| "loss": 0.6977, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00010682102628285355, | |
| "loss": 0.6966, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 0.00010494367959949937, | |
| "loss": 0.6868, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 0.00010306633291614518, | |
| "loss": 0.6802, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 0.00010118898623279097, | |
| "loss": 0.6916, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 9.931163954943679e-05, | |
| "loss": 0.7018, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 9.74342928660826e-05, | |
| "loss": 0.6951, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 9.55569461827284e-05, | |
| "loss": 0.6906, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 9.36795994993742e-05, | |
| "loss": 0.6965, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_loss": 0.6952534317970276, | |
| "eval_runtime": 65.233, | |
| "eval_samples_per_second": 30.659, | |
| "eval_steps_per_second": 0.766, | |
| "step": 1200 | |
| } | |
| ], | |
| "max_steps": 1698, | |
| "num_train_epochs": 3, | |
| "total_flos": 5.549133471969968e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |