| { | |
| "best_metric": 0.8634598271995685, | |
| "best_model_checkpoint": "result/my-unsup-simcse-bert-base-uncased", | |
| "epoch": 1.0, | |
| "global_step": 8909, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "eval_avg_sts": 0.8421121351950032, | |
| "eval_sickr_spearman": 0.820764443190438, | |
| "eval_stsb_spearman": 0.8634598271995685, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_avg_sts": 0.8272876260858589, | |
| "eval_sickr_spearman": 0.8016297649684213, | |
| "eval_stsb_spearman": 0.8529454872032965, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "eval_avg_sts": 0.8349849313583235, | |
| "eval_sickr_spearman": 0.8131009368626829, | |
| "eval_stsb_spearman": 0.8568689258539641, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 2.831630935009541e-05, | |
| "loss": 0.0001, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_avg_sts": 0.8324840345106272, | |
| "eval_sickr_spearman": 0.8115295033169321, | |
| "eval_stsb_spearman": 0.8534385657043223, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_avg_sts": 0.8342309434678776, | |
| "eval_sickr_spearman": 0.8101102323005372, | |
| "eval_stsb_spearman": 0.858351654635218, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "eval_avg_sts": 0.8358817878428014, | |
| "eval_sickr_spearman": 0.8152456216040973, | |
| "eval_stsb_spearman": 0.8565179540815056, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_avg_sts": 0.8337641137899596, | |
| "eval_sickr_spearman": 0.8073721233008548, | |
| "eval_stsb_spearman": 0.8601561042790645, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 2.663261870019082e-05, | |
| "loss": 0.0001, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "eval_avg_sts": 0.8318439410449322, | |
| "eval_sickr_spearman": 0.811522586838325, | |
| "eval_stsb_spearman": 0.8521652952515394, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_avg_sts": 0.8304909909975411, | |
| "eval_sickr_spearman": 0.8131338381671679, | |
| "eval_stsb_spearman": 0.8478481438279143, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_avg_sts": 0.829039779377259, | |
| "eval_sickr_spearman": 0.8045460213233376, | |
| "eval_stsb_spearman": 0.8535335374311805, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_avg_sts": 0.8307787364299535, | |
| "eval_sickr_spearman": 0.8137019980660793, | |
| "eval_stsb_spearman": 0.8478554747938275, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 2.4948928050286226e-05, | |
| "loss": 0.0001, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_avg_sts": 0.831658365337354, | |
| "eval_sickr_spearman": 0.8159185373352453, | |
| "eval_stsb_spearman": 0.8473981933394628, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_avg_sts": 0.8294868187051994, | |
| "eval_sickr_spearman": 0.8144402841262847, | |
| "eval_stsb_spearman": 0.8445333532841142, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_avg_sts": 0.8268186287541337, | |
| "eval_sickr_spearman": 0.8082519570169981, | |
| "eval_stsb_spearman": 0.8453853004912691, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "eval_avg_sts": 0.8313195396147719, | |
| "eval_sickr_spearman": 0.8177095210456699, | |
| "eval_stsb_spearman": 0.844929558183874, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 2.3265237400381636e-05, | |
| "loss": 0.0001, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_avg_sts": 0.8266715243566689, | |
| "eval_sickr_spearman": 0.8150033527284436, | |
| "eval_stsb_spearman": 0.8383396959848942, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_avg_sts": 0.8293624623711227, | |
| "eval_sickr_spearman": 0.8175795488851783, | |
| "eval_stsb_spearman": 0.841145375857067, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_avg_sts": 0.8322036045569485, | |
| "eval_sickr_spearman": 0.8175970802372032, | |
| "eval_stsb_spearman": 0.8468101288766937, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_avg_sts": 0.8314845597378098, | |
| "eval_sickr_spearman": 0.8156567678324077, | |
| "eval_stsb_spearman": 0.8473123516432118, | |
| "step": 2375 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 2.1581546750477045e-05, | |
| "loss": 0.0001, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_avg_sts": 0.8298162063992072, | |
| "eval_sickr_spearman": 0.8121817176433601, | |
| "eval_stsb_spearman": 0.8474506951550543, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_avg_sts": 0.8288600170060807, | |
| "eval_sickr_spearman": 0.8109708535761053, | |
| "eval_stsb_spearman": 0.8467491804360561, | |
| "step": 2625 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_avg_sts": 0.8307741487193253, | |
| "eval_sickr_spearman": 0.8109006321058029, | |
| "eval_stsb_spearman": 0.8506476653328477, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_avg_sts": 0.8299073386983135, | |
| "eval_sickr_spearman": 0.8106493814141802, | |
| "eval_stsb_spearman": 0.8491652959824467, | |
| "step": 2875 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.9897856100572458e-05, | |
| "loss": 0.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_avg_sts": 0.82947527087573, | |
| "eval_sickr_spearman": 0.808709645382602, | |
| "eval_stsb_spearman": 0.8502408963688581, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_avg_sts": 0.8247286673217802, | |
| "eval_sickr_spearman": 0.8044403048690724, | |
| "eval_stsb_spearman": 0.8450170297744881, | |
| "step": 3125 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_avg_sts": 0.8219332102471997, | |
| "eval_sickr_spearman": 0.806001651883521, | |
| "eval_stsb_spearman": 0.8378647686108784, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_avg_sts": 0.8111430899888176, | |
| "eval_sickr_spearman": 0.7991139439061943, | |
| "eval_stsb_spearman": 0.8231722360714409, | |
| "step": 3375 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.8214165450667864e-05, | |
| "loss": 0.0001, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_avg_sts": 0.807692571896139, | |
| "eval_sickr_spearman": 0.7927039052326695, | |
| "eval_stsb_spearman": 0.8226812385596084, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_avg_sts": 0.8135397552735281, | |
| "eval_sickr_spearman": 0.7942567027110624, | |
| "eval_stsb_spearman": 0.8328228078359937, | |
| "step": 3625 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_avg_sts": 0.8047796899452875, | |
| "eval_sickr_spearman": 0.7930084224157872, | |
| "eval_stsb_spearman": 0.8165509574747878, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_avg_sts": 0.8132531030215364, | |
| "eval_sickr_spearman": 0.7956294796212651, | |
| "eval_stsb_spearman": 0.8308767264218079, | |
| "step": 3875 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 1.6530474800763274e-05, | |
| "loss": 0.0001, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "eval_avg_sts": 0.8062306897971802, | |
| "eval_sickr_spearman": 0.7966211297415562, | |
| "eval_stsb_spearman": 0.8158402498528041, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "eval_avg_sts": 0.8163206710413518, | |
| "eval_sickr_spearman": 0.80029277122879, | |
| "eval_stsb_spearman": 0.8323485708539136, | |
| "step": 4125 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_avg_sts": 0.8165420284081878, | |
| "eval_sickr_spearman": 0.8011630467557471, | |
| "eval_stsb_spearman": 0.8319210100606285, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "eval_avg_sts": 0.8187419551614288, | |
| "eval_sickr_spearman": 0.8018022926847871, | |
| "eval_stsb_spearman": 0.8356816176380705, | |
| "step": 4375 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 1.4846784150858683e-05, | |
| "loss": 0.0, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_avg_sts": 0.8195743034210232, | |
| "eval_sickr_spearman": 0.8021342836579273, | |
| "eval_stsb_spearman": 0.8370143231841193, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_avg_sts": 0.8193325419945161, | |
| "eval_sickr_spearman": 0.8013091573663218, | |
| "eval_stsb_spearman": 0.8373559266227103, | |
| "step": 4625 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_avg_sts": 0.8183510820774517, | |
| "eval_sickr_spearman": 0.7999737966841394, | |
| "eval_stsb_spearman": 0.836728367470764, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "eval_avg_sts": 0.8168697079122389, | |
| "eval_sickr_spearman": 0.7952836076598091, | |
| "eval_stsb_spearman": 0.8384558081646686, | |
| "step": 4875 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 1.3163093500954093e-05, | |
| "loss": 0.0001, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "eval_avg_sts": 0.8170635469307033, | |
| "eval_sickr_spearman": 0.796297256024559, | |
| "eval_stsb_spearman": 0.8378298378368476, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_avg_sts": 0.8188600980624421, | |
| "eval_sickr_spearman": 0.7980452999302978, | |
| "eval_stsb_spearman": 0.8396748961945864, | |
| "step": 5125 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_avg_sts": 0.8179261372960408, | |
| "eval_sickr_spearman": 0.7982228709123145, | |
| "eval_stsb_spearman": 0.8376294036797671, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_avg_sts": 0.8152302038847802, | |
| "eval_sickr_spearman": 0.8007572319796964, | |
| "eval_stsb_spearman": 0.829703175789864, | |
| "step": 5375 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.14794028510495e-05, | |
| "loss": 0.0, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_avg_sts": 0.8173976618782237, | |
| "eval_sickr_spearman": 0.8015423483638039, | |
| "eval_stsb_spearman": 0.8332529753926434, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_avg_sts": 0.8175197579562594, | |
| "eval_sickr_spearman": 0.8017781330407636, | |
| "eval_stsb_spearman": 0.8332613828717551, | |
| "step": 5625 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_avg_sts": 0.8155439357729999, | |
| "eval_sickr_spearman": 0.8033175778729565, | |
| "eval_stsb_spearman": 0.8277702936730432, | |
| "step": 5750 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_avg_sts": 0.816245667230588, | |
| "eval_sickr_spearman": 0.8031514863241836, | |
| "eval_stsb_spearman": 0.8293398481369924, | |
| "step": 5875 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 9.79571220114491e-06, | |
| "loss": 0.0, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_avg_sts": 0.8173247407144557, | |
| "eval_sickr_spearman": 0.802800715190381, | |
| "eval_stsb_spearman": 0.8318487662385303, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_avg_sts": 0.8175691152062283, | |
| "eval_sickr_spearman": 0.8076647768019173, | |
| "eval_stsb_spearman": 0.8274734536105394, | |
| "step": 6125 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_avg_sts": 0.8185213107446343, | |
| "eval_sickr_spearman": 0.805431666802755, | |
| "eval_stsb_spearman": 0.8316109546865137, | |
| "step": 6250 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_avg_sts": 0.8186800157417145, | |
| "eval_sickr_spearman": 0.8050108663230556, | |
| "eval_stsb_spearman": 0.8323491651603734, | |
| "step": 6375 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 8.112021551240318e-06, | |
| "loss": 0.0001, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_avg_sts": 0.8188678392487967, | |
| "eval_sickr_spearman": 0.8050220575696909, | |
| "eval_stsb_spearman": 0.8327136209279026, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "eval_avg_sts": 0.8176842353525476, | |
| "eval_sickr_spearman": 0.805601264621933, | |
| "eval_stsb_spearman": 0.8297672060831622, | |
| "step": 6625 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_avg_sts": 0.8176889108605423, | |
| "eval_sickr_spearman": 0.8056947811764331, | |
| "eval_stsb_spearman": 0.8296830405446516, | |
| "step": 6750 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "eval_avg_sts": 0.8196884531122979, | |
| "eval_sickr_spearman": 0.8050268606798346, | |
| "eval_stsb_spearman": 0.8343500455447611, | |
| "step": 6875 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 6.428330901335728e-06, | |
| "loss": 0.0001, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_avg_sts": 0.8202633038048985, | |
| "eval_sickr_spearman": 0.8042502938317833, | |
| "eval_stsb_spearman": 0.8362763137780139, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_avg_sts": 0.819155219728974, | |
| "eval_sickr_spearman": 0.8032414485771772, | |
| "eval_stsb_spearman": 0.8350689908807707, | |
| "step": 7125 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "eval_avg_sts": 0.8185946178321757, | |
| "eval_sickr_spearman": 0.8011783206460045, | |
| "eval_stsb_spearman": 0.836010915018347, | |
| "step": 7250 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "eval_avg_sts": 0.8201412949742906, | |
| "eval_sickr_spearman": 0.8025524424270474, | |
| "eval_stsb_spearman": 0.8377301475215336, | |
| "step": 7375 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "learning_rate": 4.744640251431137e-06, | |
| "loss": 0.0, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "eval_avg_sts": 0.8212551171670207, | |
| "eval_sickr_spearman": 0.8034978385966538, | |
| "eval_stsb_spearman": 0.8390123957373876, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_avg_sts": 0.818850668226643, | |
| "eval_sickr_spearman": 0.8027982175731062, | |
| "eval_stsb_spearman": 0.8349031188801797, | |
| "step": 7625 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_avg_sts": 0.8184223281084055, | |
| "eval_sickr_spearman": 0.8041640299736004, | |
| "eval_stsb_spearman": 0.8326806262432108, | |
| "step": 7750 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_avg_sts": 0.818812249944629, | |
| "eval_sickr_spearman": 0.8036731040858015, | |
| "eval_stsb_spearman": 0.8339513958034565, | |
| "step": 7875 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 3.0609496015265466e-06, | |
| "loss": 0.0, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_avg_sts": 0.8204302396512207, | |
| "eval_sickr_spearman": 0.8052288794924833, | |
| "eval_stsb_spearman": 0.8356315998099582, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "eval_avg_sts": 0.8209129013628449, | |
| "eval_sickr_spearman": 0.8057768663287909, | |
| "eval_stsb_spearman": 0.8360489363968989, | |
| "step": 8125 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_avg_sts": 0.8191227480526566, | |
| "eval_sickr_spearman": 0.8051447290027637, | |
| "eval_stsb_spearman": 0.8331007671025495, | |
| "step": 8250 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_avg_sts": 0.8188352393246607, | |
| "eval_sickr_spearman": 0.804106104465266, | |
| "eval_stsb_spearman": 0.8335643741840555, | |
| "step": 8375 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 1.3772589516219555e-06, | |
| "loss": 0.0001, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_avg_sts": 0.8199749779902297, | |
| "eval_sickr_spearman": 0.8036332863027092, | |
| "eval_stsb_spearman": 0.8363166696777501, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_avg_sts": 0.8199803565006372, | |
| "eval_sickr_spearman": 0.8035859756677927, | |
| "eval_stsb_spearman": 0.8363747373334816, | |
| "step": 8625 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_avg_sts": 0.8201168764945796, | |
| "eval_sickr_spearman": 0.803825650863969, | |
| "eval_stsb_spearman": 0.8364081021251903, | |
| "step": 8750 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_avg_sts": 0.820079455438296, | |
| "eval_sickr_spearman": 0.8038641237762209, | |
| "eval_stsb_spearman": 0.8362947871003711, | |
| "step": 8875 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 8909, | |
| "train_runtime": 4791.076, | |
| "train_samples_per_second": 1.859 | |
| } | |
| ], | |
| "max_steps": 8909, | |
| "num_train_epochs": 1, | |
| "total_flos": 24044429181321216, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |