| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 6.802228412256268, |
| "eval_steps": 30, |
| "global_step": 300, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.22284122562674094, |
| "grad_norm": 0.2863571047782898, |
| "learning_rate": 2.785515320334262e-07, |
| "loss": 0.0285, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.4456824512534819, |
| "grad_norm": 0.2500404119491577, |
| "learning_rate": 5.571030640668524e-07, |
| "loss": 0.0396, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.6685236768802229, |
| "grad_norm": 0.3944888710975647, |
| "learning_rate": 8.356545961002786e-07, |
| "loss": 0.0396, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.6685236768802229, |
| "eval_loss": 0.037624359130859375, |
| "eval_runtime": 8.938, |
| "eval_samples_per_second": 167.823, |
| "eval_steps_per_second": 10.517, |
| "eval_sts_dev_pearson_cosine": 0.764955664513056, |
| "eval_sts_dev_pearson_dot": 0.589803108822307, |
| "eval_sts_dev_pearson_euclidean": 0.7219234025187748, |
| "eval_sts_dev_pearson_manhattan": 0.72145381158843, |
| "eval_sts_dev_pearson_max": 0.764955664513056, |
| "eval_sts_dev_spearman_cosine": 0.7647154884722602, |
| "eval_sts_dev_spearman_dot": 0.5877288570418023, |
| "eval_sts_dev_spearman_euclidean": 0.714094450799924, |
| "eval_sts_dev_spearman_manhattan": 0.7137187115899585, |
| "eval_sts_dev_spearman_max": 0.7647154884722602, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.8913649025069638, |
| "grad_norm": 0.5294222831726074, |
| "learning_rate": 1.1142061281337048e-06, |
| "loss": 0.0594, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.1337047353760445, |
| "grad_norm": 0.2601313292980194, |
| "learning_rate": 1.392757660167131e-06, |
| "loss": 0.0438, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.3565459610027855, |
| "grad_norm": 0.3371785879135132, |
| "learning_rate": 1.6713091922005572e-06, |
| "loss": 0.0302, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.3565459610027855, |
| "eval_loss": 0.03583640605211258, |
| "eval_runtime": 8.8794, |
| "eval_samples_per_second": 168.931, |
| "eval_steps_per_second": 10.586, |
| "eval_sts_dev_pearson_cosine": 0.7705187018593216, |
| "eval_sts_dev_pearson_dot": 0.6076573663583646, |
| "eval_sts_dev_pearson_euclidean": 0.7228065665910366, |
| "eval_sts_dev_pearson_manhattan": 0.7223378022429527, |
| "eval_sts_dev_pearson_max": 0.7705187018593216, |
| "eval_sts_dev_spearman_cosine": 0.7723120443468319, |
| "eval_sts_dev_spearman_dot": 0.6075378373677961, |
| "eval_sts_dev_spearman_euclidean": 0.7150378636996941, |
| "eval_sts_dev_spearman_manhattan": 0.7147730157418715, |
| "eval_sts_dev_spearman_max": 0.7723120443468319, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.5793871866295266, |
| "grad_norm": 0.3295373022556305, |
| "learning_rate": 1.9498607242339835e-06, |
| "loss": 0.0398, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.8022284122562673, |
| "grad_norm": 0.35398605465888977, |
| "learning_rate": 2.2284122562674097e-06, |
| "loss": 0.0457, |
| "step": 80 |
| }, |
| { |
| "epoch": 2.0445682451253484, |
| "grad_norm": 0.2262781709432602, |
| "learning_rate": 2.506963788300836e-06, |
| "loss": 0.0464, |
| "step": 90 |
| }, |
| { |
| "epoch": 2.0445682451253484, |
| "eval_loss": 0.03468449041247368, |
| "eval_runtime": 8.8139, |
| "eval_samples_per_second": 170.185, |
| "eval_steps_per_second": 10.665, |
| "eval_sts_dev_pearson_cosine": 0.7745596534667405, |
| "eval_sts_dev_pearson_dot": 0.6202842788366623, |
| "eval_sts_dev_pearson_euclidean": 0.7230641708414006, |
| "eval_sts_dev_pearson_manhattan": 0.7225653526096391, |
| "eval_sts_dev_pearson_max": 0.7745596534667405, |
| "eval_sts_dev_spearman_cosine": 0.7804674349325152, |
| "eval_sts_dev_spearman_dot": 0.6269246663218206, |
| "eval_sts_dev_spearman_euclidean": 0.7154276381242246, |
| "eval_sts_dev_spearman_manhattan": 0.715244268630065, |
| "eval_sts_dev_spearman_max": 0.7804674349325152, |
| "step": 90 |
| }, |
| { |
| "epoch": 2.267409470752089, |
| "grad_norm": 0.2479257434606552, |
| "learning_rate": 2.785515320334262e-06, |
| "loss": 0.026, |
| "step": 100 |
| }, |
| { |
| "epoch": 2.4902506963788302, |
| "grad_norm": 0.18173329532146454, |
| "learning_rate": 3.064066852367688e-06, |
| "loss": 0.0331, |
| "step": 110 |
| }, |
| { |
| "epoch": 2.713091922005571, |
| "grad_norm": 0.34261173009872437, |
| "learning_rate": 3.3426183844011143e-06, |
| "loss": 0.0318, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.713091922005571, |
| "eval_loss": 0.032896921038627625, |
| "eval_runtime": 8.9155, |
| "eval_samples_per_second": 168.247, |
| "eval_steps_per_second": 10.543, |
| "eval_sts_dev_pearson_cosine": 0.7787682019715072, |
| "eval_sts_dev_pearson_dot": 0.6287420367838928, |
| "eval_sts_dev_pearson_euclidean": 0.7232103414826321, |
| "eval_sts_dev_pearson_manhattan": 0.7226983341404614, |
| "eval_sts_dev_pearson_max": 0.7787682019715072, |
| "eval_sts_dev_spearman_cosine": 0.7836888215928643, |
| "eval_sts_dev_spearman_dot": 0.6396485003600499, |
| "eval_sts_dev_spearman_euclidean": 0.7152705156495667, |
| "eval_sts_dev_spearman_manhattan": 0.7150050400356657, |
| "eval_sts_dev_spearman_max": 0.7836888215928643, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.935933147632312, |
| "grad_norm": 0.32490110397338867, |
| "learning_rate": 3.6211699164345405e-06, |
| "loss": 0.0399, |
| "step": 130 |
| }, |
| { |
| "epoch": 3.1782729805013927, |
| "grad_norm": 0.2102952003479004, |
| "learning_rate": 3.899721448467967e-06, |
| "loss": 0.0264, |
| "step": 140 |
| }, |
| { |
| "epoch": 3.401114206128134, |
| "grad_norm": 0.29396915435791016, |
| "learning_rate": 4.178272980501394e-06, |
| "loss": 0.0268, |
| "step": 150 |
| }, |
| { |
| "epoch": 3.401114206128134, |
| "eval_loss": 0.03317762911319733, |
| "eval_runtime": 9.3558, |
| "eval_samples_per_second": 160.329, |
| "eval_steps_per_second": 10.047, |
| "eval_sts_dev_pearson_cosine": 0.7818829552806165, |
| "eval_sts_dev_pearson_dot": 0.6407030863223679, |
| "eval_sts_dev_pearson_euclidean": 0.7251646010881555, |
| "eval_sts_dev_pearson_manhattan": 0.724664700544263, |
| "eval_sts_dev_pearson_max": 0.7818829552806165, |
| "eval_sts_dev_spearman_cosine": 0.7883531423647104, |
| "eval_sts_dev_spearman_dot": 0.6500865416788887, |
| "eval_sts_dev_spearman_euclidean": 0.7177286550860219, |
| "eval_sts_dev_spearman_manhattan": 0.717444216928271, |
| "eval_sts_dev_spearman_max": 0.7883531423647104, |
| "step": 150 |
| }, |
| { |
| "epoch": 3.6239554317548746, |
| "grad_norm": 0.18695248663425446, |
| "learning_rate": 4.456824512534819e-06, |
| "loss": 0.0241, |
| "step": 160 |
| }, |
| { |
| "epoch": 3.8467966573816157, |
| "grad_norm": 0.2890850007534027, |
| "learning_rate": 4.735376044568246e-06, |
| "loss": 0.0309, |
| "step": 170 |
| }, |
| { |
| "epoch": 4.089136490250697, |
| "grad_norm": 0.1940670907497406, |
| "learning_rate": 5.013927576601672e-06, |
| "loss": 0.0263, |
| "step": 180 |
| }, |
| { |
| "epoch": 4.089136490250697, |
| "eval_loss": 0.03255865350365639, |
| "eval_runtime": 9.7386, |
| "eval_samples_per_second": 154.026, |
| "eval_steps_per_second": 9.652, |
| "eval_sts_dev_pearson_cosine": 0.7866026930184947, |
| "eval_sts_dev_pearson_dot": 0.6395121549516076, |
| "eval_sts_dev_pearson_euclidean": 0.7276962342001664, |
| "eval_sts_dev_pearson_manhattan": 0.7271401753294904, |
| "eval_sts_dev_pearson_max": 0.7866026930184947, |
| "eval_sts_dev_spearman_cosine": 0.7917956870568429, |
| "eval_sts_dev_spearman_dot": 0.6487597753154516, |
| "eval_sts_dev_spearman_euclidean": 0.7209760720407495, |
| "eval_sts_dev_spearman_manhattan": 0.720422142644627, |
| "eval_sts_dev_spearman_max": 0.7917956870568429, |
| "step": 180 |
| }, |
| { |
| "epoch": 4.311977715877437, |
| "grad_norm": 0.18042121827602386, |
| "learning_rate": 5.292479108635098e-06, |
| "loss": 0.0164, |
| "step": 190 |
| }, |
| { |
| "epoch": 4.534818941504178, |
| "grad_norm": 0.15695632994174957, |
| "learning_rate": 5.571030640668524e-06, |
| "loss": 0.0226, |
| "step": 200 |
| }, |
| { |
| "epoch": 4.757660167130919, |
| "grad_norm": 0.22100669145584106, |
| "learning_rate": 5.849582172701951e-06, |
| "loss": 0.0196, |
| "step": 210 |
| }, |
| { |
| "epoch": 4.757660167130919, |
| "eval_loss": 0.03137701749801636, |
| "eval_runtime": 9.3122, |
| "eval_samples_per_second": 161.08, |
| "eval_steps_per_second": 10.094, |
| "eval_sts_dev_pearson_cosine": 0.7870442789973777, |
| "eval_sts_dev_pearson_dot": 0.6305251777821479, |
| "eval_sts_dev_pearson_euclidean": 0.7283553144330797, |
| "eval_sts_dev_pearson_manhattan": 0.7277896396418712, |
| "eval_sts_dev_pearson_max": 0.7870442789973777, |
| "eval_sts_dev_spearman_cosine": 0.7896056186150787, |
| "eval_sts_dev_spearman_dot": 0.6435332170567882, |
| "eval_sts_dev_spearman_euclidean": 0.7218272927999315, |
| "eval_sts_dev_spearman_manhattan": 0.7214883882970341, |
| "eval_sts_dev_spearman_max": 0.7896056186150787, |
| "step": 210 |
| }, |
| { |
| "epoch": 4.9805013927576605, |
| "grad_norm": 0.23286058008670807, |
| "learning_rate": 6.128133704735376e-06, |
| "loss": 0.0217, |
| "step": 220 |
| }, |
| { |
| "epoch": 5.222841225626741, |
| "grad_norm": 0.17037726938724518, |
| "learning_rate": 6.406685236768803e-06, |
| "loss": 0.0134, |
| "step": 230 |
| }, |
| { |
| "epoch": 5.445682451253482, |
| "grad_norm": 0.18465109169483185, |
| "learning_rate": 6.685236768802229e-06, |
| "loss": 0.0157, |
| "step": 240 |
| }, |
| { |
| "epoch": 5.445682451253482, |
| "eval_loss": 0.03204584866762161, |
| "eval_runtime": 9.2367, |
| "eval_samples_per_second": 162.396, |
| "eval_steps_per_second": 10.177, |
| "eval_sts_dev_pearson_cosine": 0.7879215371135696, |
| "eval_sts_dev_pearson_dot": 0.6400228300721227, |
| "eval_sts_dev_pearson_euclidean": 0.7281787730138891, |
| "eval_sts_dev_pearson_manhattan": 0.7275982374933561, |
| "eval_sts_dev_pearson_max": 0.7879215371135696, |
| "eval_sts_dev_spearman_cosine": 0.7910741437085514, |
| "eval_sts_dev_spearman_dot": 0.6477074504092454, |
| "eval_sts_dev_spearman_euclidean": 0.7215931954085069, |
| "eval_sts_dev_spearman_manhattan": 0.7212215794866459, |
| "eval_sts_dev_spearman_max": 0.7910741437085514, |
| "step": 240 |
| }, |
| { |
| "epoch": 5.6685236768802225, |
| "grad_norm": 0.2040845900774002, |
| "learning_rate": 6.963788300835655e-06, |
| "loss": 0.0136, |
| "step": 250 |
| }, |
| { |
| "epoch": 5.891364902506964, |
| "grad_norm": 0.2258893996477127, |
| "learning_rate": 7.242339832869081e-06, |
| "loss": 0.0143, |
| "step": 260 |
| }, |
| { |
| "epoch": 6.133704735376044, |
| "grad_norm": 0.11912677437067032, |
| "learning_rate": 7.5208913649025075e-06, |
| "loss": 0.0114, |
| "step": 270 |
| }, |
| { |
| "epoch": 6.133704735376044, |
| "eval_loss": 0.03218431398272514, |
| "eval_runtime": 9.1387, |
| "eval_samples_per_second": 164.137, |
| "eval_steps_per_second": 10.286, |
| "eval_sts_dev_pearson_cosine": 0.7875077785532312, |
| "eval_sts_dev_pearson_dot": 0.6389110741592561, |
| "eval_sts_dev_pearson_euclidean": 0.7278472242426817, |
| "eval_sts_dev_pearson_manhattan": 0.7272084663171942, |
| "eval_sts_dev_pearson_max": 0.7875077785532312, |
| "eval_sts_dev_spearman_cosine": 0.7906616591294269, |
| "eval_sts_dev_spearman_dot": 0.6460173261770129, |
| "eval_sts_dev_spearman_euclidean": 0.721550246781466, |
| "eval_sts_dev_spearman_manhattan": 0.7213414755418661, |
| "eval_sts_dev_spearman_max": 0.7906616591294269, |
| "step": 270 |
| }, |
| { |
| "epoch": 6.3565459610027855, |
| "grad_norm": 0.151308074593544, |
| "learning_rate": 7.799442896935934e-06, |
| "loss": 0.0077, |
| "step": 280 |
| }, |
| { |
| "epoch": 6.579387186629527, |
| "grad_norm": 0.12630996108055115, |
| "learning_rate": 8.07799442896936e-06, |
| "loss": 0.0116, |
| "step": 290 |
| }, |
| { |
| "epoch": 6.802228412256268, |
| "grad_norm": 0.13274052739143372, |
| "learning_rate": 8.356545961002787e-06, |
| "loss": 0.0087, |
| "step": 300 |
| }, |
| { |
| "epoch": 6.802228412256268, |
| "eval_loss": 0.03128228709101677, |
| "eval_runtime": 9.1866, |
| "eval_samples_per_second": 163.282, |
| "eval_steps_per_second": 10.232, |
| "eval_sts_dev_pearson_cosine": 0.7861496812877409, |
| "eval_sts_dev_pearson_dot": 0.6275374593246463, |
| "eval_sts_dev_pearson_euclidean": 0.7262636157490319, |
| "eval_sts_dev_pearson_manhattan": 0.7255528346187912, |
| "eval_sts_dev_pearson_max": 0.7861496812877409, |
| "eval_sts_dev_spearman_cosine": 0.7868100521872967, |
| "eval_sts_dev_spearman_dot": 0.6389952669242227, |
| "eval_sts_dev_spearman_euclidean": 0.7203939324818445, |
| "eval_sts_dev_spearman_manhattan": 0.7200129773124266, |
| "eval_sts_dev_spearman_max": 0.7868100521872967, |
| "step": 300 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 440, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 10, |
| "save_steps": 300, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
