| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 500, | |
| "global_step": 6313, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0792016473942658, | |
| "grad_norm": 0.07919532805681229, | |
| "learning_rate": 3.9556962025316456e-05, | |
| "loss": 0.1311, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.0792016473942658, | |
| "eval_loss": 0.014126550406217575, | |
| "eval_runtime": 83.0348, | |
| "eval_samples_per_second": 60.216, | |
| "eval_steps_per_second": 1.891, | |
| "eval_sts-dev_pearson_cosine": 0.1557480599999687, | |
| "eval_sts-dev_pearson_dot": 0.13799264585261115, | |
| "eval_sts-dev_pearson_euclidean": 0.18714752476831273, | |
| "eval_sts-dev_pearson_manhattan": 0.18855989459573652, | |
| "eval_sts-dev_pearson_max": 0.18855989459573652, | |
| "eval_sts-dev_spearman_cosine": 0.20357424121861373, | |
| "eval_sts-dev_spearman_dot": 0.14034776792350498, | |
| "eval_sts-dev_spearman_euclidean": 0.21044421396782537, | |
| "eval_sts-dev_spearman_manhattan": 0.21183531991746804, | |
| "eval_sts-dev_spearman_max": 0.21183531991746804, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.1584032947885316, | |
| "grad_norm": 0.012879022397100925, | |
| "learning_rate": 4.677873613800388e-05, | |
| "loss": 0.0203, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.1584032947885316, | |
| "eval_loss": 0.015841394662857056, | |
| "eval_runtime": 82.5774, | |
| "eval_samples_per_second": 60.549, | |
| "eval_steps_per_second": 1.901, | |
| "eval_sts-dev_pearson_cosine": 0.16793876783774, | |
| "eval_sts-dev_pearson_dot": 0.1481957278796634, | |
| "eval_sts-dev_pearson_euclidean": 0.19722850814365245, | |
| "eval_sts-dev_pearson_manhattan": 0.19617097331605537, | |
| "eval_sts-dev_pearson_max": 0.19722850814365245, | |
| "eval_sts-dev_spearman_cosine": 0.19966948583403588, | |
| "eval_sts-dev_spearman_dot": 0.14869572921537724, | |
| "eval_sts-dev_spearman_euclidean": 0.208144651870388, | |
| "eval_sts-dev_spearman_manhattan": 0.20697988059772135, | |
| "eval_sts-dev_spearman_max": 0.208144651870388, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.2376049421827974, | |
| "grad_norm": 0.007155057042837143, | |
| "learning_rate": 4.2378102446752334e-05, | |
| "loss": 0.0174, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.2376049421827974, | |
| "eval_loss": 0.01738722249865532, | |
| "eval_runtime": 82.5636, | |
| "eval_samples_per_second": 60.559, | |
| "eval_steps_per_second": 1.902, | |
| "eval_sts-dev_pearson_cosine": 0.13663702381510662, | |
| "eval_sts-dev_pearson_dot": 0.11588852598989118, | |
| "eval_sts-dev_pearson_euclidean": 0.16056667152911408, | |
| "eval_sts-dev_pearson_manhattan": 0.15924268782656817, | |
| "eval_sts-dev_pearson_max": 0.16056667152911408, | |
| "eval_sts-dev_spearman_cosine": 0.16527466832006016, | |
| "eval_sts-dev_spearman_dot": 0.10078185264794931, | |
| "eval_sts-dev_spearman_euclidean": 0.1719215715191216, | |
| "eval_sts-dev_spearman_manhattan": 0.170303502778187, | |
| "eval_sts-dev_spearman_max": 0.1719215715191216, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.3168065895770632, | |
| "grad_norm": 0.002368535613641143, | |
| "learning_rate": 3.79774687555008e-05, | |
| "loss": 0.0108, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.3168065895770632, | |
| "eval_loss": 0.013646911829710007, | |
| "eval_runtime": 83.1873, | |
| "eval_samples_per_second": 60.105, | |
| "eval_steps_per_second": 1.887, | |
| "eval_sts-dev_pearson_cosine": 0.14207290429771552, | |
| "eval_sts-dev_pearson_dot": 0.12219302310028649, | |
| "eval_sts-dev_pearson_euclidean": 0.15249164642806468, | |
| "eval_sts-dev_pearson_manhattan": 0.15099460310457263, | |
| "eval_sts-dev_pearson_max": 0.15249164642806468, | |
| "eval_sts-dev_spearman_cosine": 0.14567413155731176, | |
| "eval_sts-dev_spearman_dot": 0.09939489795167657, | |
| "eval_sts-dev_spearman_euclidean": 0.14960035136962835, | |
| "eval_sts-dev_spearman_manhattan": 0.14872808741782187, | |
| "eval_sts-dev_spearman_max": 0.14960035136962835, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.39600823697132903, | |
| "grad_norm": 0.011403551325201988, | |
| "learning_rate": 3.3576835064249254e-05, | |
| "loss": 0.0121, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.39600823697132903, | |
| "eval_loss": 0.015611983835697174, | |
| "eval_runtime": 82.8829, | |
| "eval_samples_per_second": 60.326, | |
| "eval_steps_per_second": 1.894, | |
| "eval_sts-dev_pearson_cosine": 0.1786266334873075, | |
| "eval_sts-dev_pearson_dot": 0.16571459472812308, | |
| "eval_sts-dev_pearson_euclidean": 0.19586188718253267, | |
| "eval_sts-dev_pearson_manhattan": 0.19603778309890557, | |
| "eval_sts-dev_pearson_max": 0.19603778309890557, | |
| "eval_sts-dev_spearman_cosine": 0.20990140220242978, | |
| "eval_sts-dev_spearman_dot": 0.1668269410484095, | |
| "eval_sts-dev_spearman_euclidean": 0.21443661354146873, | |
| "eval_sts-dev_spearman_manhattan": 0.21529338637929912, | |
| "eval_sts-dev_spearman_max": 0.21529338637929912, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.4752098843655948, | |
| "grad_norm": 0.004168146755546331, | |
| "learning_rate": 2.9176201372997714e-05, | |
| "loss": 0.0122, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.4752098843655948, | |
| "eval_loss": 0.013952625915408134, | |
| "eval_runtime": 82.7861, | |
| "eval_samples_per_second": 60.397, | |
| "eval_steps_per_second": 1.896, | |
| "eval_sts-dev_pearson_cosine": 0.16089480397614714, | |
| "eval_sts-dev_pearson_dot": 0.14333252395560012, | |
| "eval_sts-dev_pearson_euclidean": 0.17458568236482797, | |
| "eval_sts-dev_pearson_manhattan": 0.1736800184097837, | |
| "eval_sts-dev_pearson_max": 0.17458568236482797, | |
| "eval_sts-dev_spearman_cosine": 0.17227083866593193, | |
| "eval_sts-dev_spearman_dot": 0.12781779851368713, | |
| "eval_sts-dev_spearman_euclidean": 0.17703810710585532, | |
| "eval_sts-dev_spearman_manhattan": 0.17557253669161538, | |
| "eval_sts-dev_spearman_max": 0.17703810710585532, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.5544115317598606, | |
| "grad_norm": 0.004282405134290457, | |
| "learning_rate": 2.4775567681746174e-05, | |
| "loss": 0.0125, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5544115317598606, | |
| "eval_loss": 0.011783541180193424, | |
| "eval_runtime": 82.8159, | |
| "eval_samples_per_second": 60.375, | |
| "eval_steps_per_second": 1.896, | |
| "eval_sts-dev_pearson_cosine": 0.18448919166260044, | |
| "eval_sts-dev_pearson_dot": 0.17905699568214264, | |
| "eval_sts-dev_pearson_euclidean": 0.20624063360858977, | |
| "eval_sts-dev_pearson_manhattan": 0.205619351099398, | |
| "eval_sts-dev_pearson_max": 0.20624063360858977, | |
| "eval_sts-dev_spearman_cosine": 0.22476168122019577, | |
| "eval_sts-dev_spearman_dot": 0.19305224567026316, | |
| "eval_sts-dev_spearman_euclidean": 0.2268539691521332, | |
| "eval_sts-dev_spearman_manhattan": 0.2268021523901189, | |
| "eval_sts-dev_spearman_max": 0.2268539691521332, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.6336131791541264, | |
| "grad_norm": 0.016134686768054962, | |
| "learning_rate": 2.0374933990494634e-05, | |
| "loss": 0.0079, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.6336131791541264, | |
| "eval_loss": 0.011526196263730526, | |
| "eval_runtime": 83.1812, | |
| "eval_samples_per_second": 60.11, | |
| "eval_steps_per_second": 1.887, | |
| "eval_sts-dev_pearson_cosine": 0.16979017817169434, | |
| "eval_sts-dev_pearson_dot": 0.16329193511035556, | |
| "eval_sts-dev_pearson_euclidean": 0.2041557241070686, | |
| "eval_sts-dev_pearson_manhattan": 0.20415667390135622, | |
| "eval_sts-dev_pearson_max": 0.20415667390135622, | |
| "eval_sts-dev_spearman_cosine": 0.23370816253094054, | |
| "eval_sts-dev_spearman_dot": 0.19110938133669397, | |
| "eval_sts-dev_spearman_euclidean": 0.23731458674719166, | |
| "eval_sts-dev_spearman_manhattan": 0.2363744330684564, | |
| "eval_sts-dev_spearman_max": 0.23731458674719166, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.7128148265483922, | |
| "grad_norm": 0.10654988884925842, | |
| "learning_rate": 1.5983101566625593e-05, | |
| "loss": 0.0093, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.7128148265483922, | |
| "eval_loss": 0.01042733620852232, | |
| "eval_runtime": 82.4696, | |
| "eval_samples_per_second": 60.628, | |
| "eval_steps_per_second": 1.904, | |
| "eval_sts-dev_pearson_cosine": 0.16883832363197002, | |
| "eval_sts-dev_pearson_dot": 0.16172877878537467, | |
| "eval_sts-dev_pearson_euclidean": 0.20343671061551505, | |
| "eval_sts-dev_pearson_manhattan": 0.20269317144225543, | |
| "eval_sts-dev_pearson_max": 0.20343671061551505, | |
| "eval_sts-dev_spearman_cosine": 0.23314123864913222, | |
| "eval_sts-dev_spearman_dot": 0.19029123486315452, | |
| "eval_sts-dev_spearman_euclidean": 0.23747745874658102, | |
| "eval_sts-dev_spearman_manhattan": 0.23681371095402073, | |
| "eval_sts-dev_spearman_max": 0.23747745874658102, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.7920164739426581, | |
| "grad_norm": 0.0034745726734399796, | |
| "learning_rate": 1.1582467875374054e-05, | |
| "loss": 0.0071, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.7920164739426581, | |
| "eval_loss": 0.010719917714595795, | |
| "eval_runtime": 82.5902, | |
| "eval_samples_per_second": 60.54, | |
| "eval_steps_per_second": 1.901, | |
| "eval_sts-dev_pearson_cosine": 0.19565242771314767, | |
| "eval_sts-dev_pearson_dot": 0.18923012649171922, | |
| "eval_sts-dev_pearson_euclidean": 0.2278768429358364, | |
| "eval_sts-dev_pearson_manhattan": 0.22768897126347665, | |
| "eval_sts-dev_pearson_max": 0.2278768429358364, | |
| "eval_sts-dev_spearman_cosine": 0.24240094548214325, | |
| "eval_sts-dev_spearman_dot": 0.21284981986619678, | |
| "eval_sts-dev_spearman_euclidean": 0.24732163285243935, | |
| "eval_sts-dev_spearman_manhattan": 0.24736929658665446, | |
| "eval_sts-dev_spearman_max": 0.24736929658665446, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.8712181213369238, | |
| "grad_norm": 0.004818719811737537, | |
| "learning_rate": 7.181834184122514e-06, | |
| "loss": 0.0041, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.8712181213369238, | |
| "eval_loss": 0.009991911239922047, | |
| "eval_runtime": 82.1637, | |
| "eval_samples_per_second": 60.854, | |
| "eval_steps_per_second": 1.911, | |
| "eval_sts-dev_pearson_cosine": 0.1892693811813182, | |
| "eval_sts-dev_pearson_dot": 0.18290497740650222, | |
| "eval_sts-dev_pearson_euclidean": 0.22463164519842746, | |
| "eval_sts-dev_pearson_manhattan": 0.22460268853676083, | |
| "eval_sts-dev_pearson_max": 0.22463164519842746, | |
| "eval_sts-dev_spearman_cosine": 0.24628027091826607, | |
| "eval_sts-dev_spearman_dot": 0.21127633301239485, | |
| "eval_sts-dev_spearman_euclidean": 0.251238048387475, | |
| "eval_sts-dev_spearman_manhattan": 0.2518403299451181, | |
| "eval_sts-dev_spearman_max": 0.2518403299451181, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.9504197687311896, | |
| "grad_norm": 0.012634661048650742, | |
| "learning_rate": 2.7812004928709737e-06, | |
| "loss": 0.0069, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.9504197687311896, | |
| "eval_loss": 0.009837556630373001, | |
| "eval_runtime": 83.3155, | |
| "eval_samples_per_second": 60.013, | |
| "eval_steps_per_second": 1.884, | |
| "eval_sts-dev_pearson_cosine": 0.19373258731869963, | |
| "eval_sts-dev_pearson_dot": 0.18775862207030505, | |
| "eval_sts-dev_pearson_euclidean": 0.22537635202224982, | |
| "eval_sts-dev_pearson_manhattan": 0.2245827911400446, | |
| "eval_sts-dev_pearson_max": 0.22537635202224982, | |
| "eval_sts-dev_spearman_cosine": 0.24307341815427166, | |
| "eval_sts-dev_spearman_dot": 0.2124049530103558, | |
| "eval_sts-dev_spearman_euclidean": 0.24695143686545143, | |
| "eval_sts-dev_spearman_manhattan": 0.2468102784042943, | |
| "eval_sts-dev_spearman_max": 0.24695143686545143, | |
| "step": 6000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6313, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |