| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9307520476545048, | |
| "eval_steps": 1000, | |
| "global_step": 15000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0062050136510300325, | |
| "grad_norm": 101.88961029052734, | |
| "learning_rate": 0.0001, | |
| "loss": 4.5088, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.012410027302060065, | |
| "grad_norm": 99.7363510131836, | |
| "learning_rate": 9.999038127056248e-05, | |
| "loss": 0.8644, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.018615040953090096, | |
| "grad_norm": 59.41786575317383, | |
| "learning_rate": 9.996152878304816e-05, | |
| "loss": 0.7189, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.02482005460412013, | |
| "grad_norm": 61.05416488647461, | |
| "learning_rate": 9.991345363842789e-05, | |
| "loss": 0.6253, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.03102506825515016, | |
| "grad_norm": 53.87808609008789, | |
| "learning_rate": 9.9846174333574e-05, | |
| "loss": 0.5863, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03723008190618019, | |
| "grad_norm": 48.58293914794922, | |
| "learning_rate": 9.975971675414371e-05, | |
| "loss": 0.5555, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.04343509555721023, | |
| "grad_norm": 42.353153228759766, | |
| "learning_rate": 9.965411416461959e-05, | |
| "loss": 0.5566, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.04964010920824026, | |
| "grad_norm": 46.06748580932617, | |
| "learning_rate": 9.952940719551112e-05, | |
| "loss": 0.5425, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.05584512285927029, | |
| "grad_norm": 46.486637115478516, | |
| "learning_rate": 9.938564382772205e-05, | |
| "loss": 0.5004, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.06205013651030032, | |
| "grad_norm": 43.99421691894531, | |
| "learning_rate": 9.922287937408994e-05, | |
| "loss": 0.4722, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06205013651030032, | |
| "eval_avg_non_pair_similarity": -0.00011276310920446573, | |
| "eval_avg_pair_similarity": 0.0020758428336121143, | |
| "eval_loss": 0.44835320115089417, | |
| "eval_runtime": 52.6112, | |
| "eval_samples_per_second": 9.504, | |
| "eval_similarity_ratio": -18.408882552609725, | |
| "eval_steps_per_second": 0.304, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06825515016133035, | |
| "grad_norm": 38.14698791503906, | |
| "learning_rate": 9.904117645810441e-05, | |
| "loss": 0.4444, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.07446016381236038, | |
| "grad_norm": 29.508113861083984, | |
| "learning_rate": 9.884060498981296e-05, | |
| "loss": 0.4615, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.08066517746339041, | |
| "grad_norm": 34.64622497558594, | |
| "learning_rate": 9.862124213892304e-05, | |
| "loss": 0.4665, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.08687019111442046, | |
| "grad_norm": 40.75010299682617, | |
| "learning_rate": 9.838317230511112e-05, | |
| "loss": 0.4275, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.09307520476545049, | |
| "grad_norm": 28.284589767456055, | |
| "learning_rate": 9.81264870855499e-05, | |
| "loss": 0.4129, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.09928021841648052, | |
| "grad_norm": 34.2197151184082, | |
| "learning_rate": 9.785128523966653e-05, | |
| "loss": 0.4389, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.10548523206751055, | |
| "grad_norm": 27.808895111083984, | |
| "learning_rate": 9.755767265114484e-05, | |
| "loss": 0.4184, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.11169024571854058, | |
| "grad_norm": 35.00907897949219, | |
| "learning_rate": 9.724576228718678e-05, | |
| "loss": 0.3995, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.11789525936957061, | |
| "grad_norm": 30.57769203186035, | |
| "learning_rate": 9.691567415504832e-05, | |
| "loss": 0.415, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.12410027302060064, | |
| "grad_norm": 26.989404678344727, | |
| "learning_rate": 9.656753525586681e-05, | |
| "loss": 0.4052, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.12410027302060064, | |
| "eval_avg_non_pair_similarity": 0.0006865022985091086, | |
| "eval_avg_pair_similarity": 0.007665629971772433, | |
| "eval_loss": 0.3601702153682709, | |
| "eval_runtime": 52.565, | |
| "eval_samples_per_second": 9.512, | |
| "eval_similarity_ratio": 11.166211662247951, | |
| "eval_steps_per_second": 0.304, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.1303052866716307, | |
| "grad_norm": 32.37137222290039, | |
| "learning_rate": 9.620147953579737e-05, | |
| "loss": 0.3979, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.1365103003226607, | |
| "grad_norm": 22.6475772857666, | |
| "learning_rate": 9.581764783447719e-05, | |
| "loss": 0.3807, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.14271531397369075, | |
| "grad_norm": 22.612077713012695, | |
| "learning_rate": 9.54161878308377e-05, | |
| "loss": 0.3711, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.14892032762472077, | |
| "grad_norm": 30.973310470581055, | |
| "learning_rate": 9.499725398628507e-05, | |
| "loss": 0.3658, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.1551253412757508, | |
| "grad_norm": 24.443492889404297, | |
| "learning_rate": 9.456100748527143e-05, | |
| "loss": 0.37, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.16133035492678083, | |
| "grad_norm": 33.75666809082031, | |
| "learning_rate": 9.410761617327921e-05, | |
| "loss": 0.3512, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.16753536857781087, | |
| "grad_norm": 23.727365493774414, | |
| "learning_rate": 9.363725449224282e-05, | |
| "loss": 0.3318, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.17374038222884092, | |
| "grad_norm": 28.48720359802246, | |
| "learning_rate": 9.315010341343213e-05, | |
| "loss": 0.3729, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.17994539587987093, | |
| "grad_norm": 28.53714370727539, | |
| "learning_rate": 9.264635036782405e-05, | |
| "loss": 0.3369, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.18615040953090098, | |
| "grad_norm": 21.766740798950195, | |
| "learning_rate": 9.212618917398855e-05, | |
| "loss": 0.317, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.18615040953090098, | |
| "eval_avg_non_pair_similarity": 0.00025785160030682527, | |
| "eval_avg_pair_similarity": 0.0073178326906636355, | |
| "eval_loss": 0.323689728975296, | |
| "eval_runtime": 52.8381, | |
| "eval_samples_per_second": 9.463, | |
| "eval_similarity_ratio": 28.38001657525464, | |
| "eval_steps_per_second": 0.303, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.192355423181931, | |
| "grad_norm": 25.073217391967773, | |
| "learning_rate": 9.158981996351711e-05, | |
| "loss": 0.3092, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.19856043683296104, | |
| "grad_norm": 25.02987289428711, | |
| "learning_rate": 9.10374491040223e-05, | |
| "loss": 0.3177, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.20476545048399106, | |
| "grad_norm": 23.147781372070312, | |
| "learning_rate": 9.046928911973799e-05, | |
| "loss": 0.3357, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.2109704641350211, | |
| "grad_norm": 18.975126266479492, | |
| "learning_rate": 8.988555860975082e-05, | |
| "loss": 0.2946, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.21717547778605112, | |
| "grad_norm": 22.149898529052734, | |
| "learning_rate": 8.92864821638944e-05, | |
| "loss": 0.3184, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.22338049143708116, | |
| "grad_norm": 31.714447021484375, | |
| "learning_rate": 8.86722902763385e-05, | |
| "loss": 0.315, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.22958550508811118, | |
| "grad_norm": 22.137062072753906, | |
| "learning_rate": 8.804321925690672e-05, | |
| "loss": 0.3007, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.23579051873914122, | |
| "grad_norm": 27.34349250793457, | |
| "learning_rate": 8.739951114015636e-05, | |
| "loss": 0.3297, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.24199553239017127, | |
| "grad_norm": 20.715852737426758, | |
| "learning_rate": 8.674141359225589e-05, | |
| "loss": 0.2943, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.2482005460412013, | |
| "grad_norm": 21.26046371459961, | |
| "learning_rate": 8.60691798156956e-05, | |
| "loss": 0.284, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.2482005460412013, | |
| "eval_avg_non_pair_similarity": 0.00036512027781798257, | |
| "eval_avg_pair_similarity": 0.002839699084404856, | |
| "eval_loss": 0.26940277218818665, | |
| "eval_runtime": 52.6691, | |
| "eval_samples_per_second": 9.493, | |
| "eval_similarity_ratio": 7.777434606961175, | |
| "eval_steps_per_second": 0.304, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.25440555969223133, | |
| "grad_norm": 22.599706649780273, | |
| "learning_rate": 8.538306845186808e-05, | |
| "loss": 0.2942, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.2606105733432614, | |
| "grad_norm": 16.177858352661133, | |
| "learning_rate": 8.468334348155624e-05, | |
| "loss": 0.2874, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.26681558699429136, | |
| "grad_norm": 15.479412078857422, | |
| "learning_rate": 8.39702741233669e-05, | |
| "loss": 0.2758, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.2730206006453214, | |
| "grad_norm": 20.40460777282715, | |
| "learning_rate": 8.324413473014913e-05, | |
| "loss": 0.2788, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.27922561429635145, | |
| "grad_norm": 15.524222373962402, | |
| "learning_rate": 8.250520468343722e-05, | |
| "loss": 0.2667, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.2854306279473815, | |
| "grad_norm": 19.124486923217773, | |
| "learning_rate": 8.175376828595885e-05, | |
| "loss": 0.2828, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.29163564159841154, | |
| "grad_norm": 14.729342460632324, | |
| "learning_rate": 8.09901146522499e-05, | |
| "loss": 0.2757, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.29784065524944153, | |
| "grad_norm": 19.547435760498047, | |
| "learning_rate": 8.021453759741784e-05, | |
| "loss": 0.3102, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.3040456689004716, | |
| "grad_norm": 16.095699310302734, | |
| "learning_rate": 7.942733552409662e-05, | |
| "loss": 0.2791, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.3102506825515016, | |
| "grad_norm": 25.538352966308594, | |
| "learning_rate": 7.862881130763646e-05, | |
| "loss": 0.2487, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.3102506825515016, | |
| "eval_avg_non_pair_similarity": -2.6742616108896766e-05, | |
| "eval_avg_pair_similarity": 0.007982193220406771, | |
| "eval_loss": 0.3381979465484619, | |
| "eval_runtime": 52.9305, | |
| "eval_samples_per_second": 9.446, | |
| "eval_similarity_ratio": -298.4821375703496, | |
| "eval_steps_per_second": 0.302, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.31645569620253167, | |
| "grad_norm": 19.506053924560547, | |
| "learning_rate": 7.781927217957286e-05, | |
| "loss": 0.2534, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.32266070985356166, | |
| "grad_norm": 17.35991668701172, | |
| "learning_rate": 7.699902960941952e-05, | |
| "loss": 0.2591, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.3288657235045917, | |
| "grad_norm": 18.530628204345703, | |
| "learning_rate": 7.616839918483061e-05, | |
| "loss": 0.2504, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.33507073715562175, | |
| "grad_norm": 16.918973922729492, | |
| "learning_rate": 7.532770049017883e-05, | |
| "loss": 0.2697, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.3412757508066518, | |
| "grad_norm": 15.549245834350586, | |
| "learning_rate": 7.447725698359547e-05, | |
| "loss": 0.269, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.34748076445768183, | |
| "grad_norm": 14.331618309020996, | |
| "learning_rate": 7.361739587252019e-05, | |
| "loss": 0.2466, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.3536857781087118, | |
| "grad_norm": 13.95034122467041, | |
| "learning_rate": 7.274844798780826e-05, | |
| "loss": 0.2491, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.35989079175974187, | |
| "grad_norm": 13.919034004211426, | |
| "learning_rate": 7.18707476564436e-05, | |
| "loss": 0.2189, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.3660958054107719, | |
| "grad_norm": 15.497775077819824, | |
| "learning_rate": 7.098463257290679e-05, | |
| "loss": 0.1965, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.37230081906180196, | |
| "grad_norm": 17.710037231445312, | |
| "learning_rate": 7.009044366924739e-05, | |
| "loss": 0.2509, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.37230081906180196, | |
| "eval_avg_non_pair_similarity": 2.055158477392279e-05, | |
| "eval_avg_pair_similarity": 0.006798229112289846, | |
| "eval_loss": 0.24542327225208282, | |
| "eval_runtime": 52.6713, | |
| "eval_samples_per_second": 9.493, | |
| "eval_similarity_ratio": 330.78855898820456, | |
| "eval_steps_per_second": 0.304, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.37850583271283195, | |
| "grad_norm": 13.973624229431152, | |
| "learning_rate": 6.918852498391063e-05, | |
| "loss": 0.2343, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.384710846363862, | |
| "grad_norm": 14.253776550292969, | |
| "learning_rate": 6.827922352936886e-05, | |
| "loss": 0.2131, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.39091586001489204, | |
| "grad_norm": 22.384563446044922, | |
| "learning_rate": 6.736288915860883e-05, | |
| "loss": 0.25, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.3971208736659221, | |
| "grad_norm": 14.788619041442871, | |
| "learning_rate": 6.643987443052595e-05, | |
| "loss": 0.2296, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.40332588731695207, | |
| "grad_norm": 14.269457817077637, | |
| "learning_rate": 6.551053447427773e-05, | |
| "loss": 0.2208, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.4095309009679821, | |
| "grad_norm": 14.065217971801758, | |
| "learning_rate": 6.457522685264792e-05, | |
| "loss": 0.2262, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.41573591461901216, | |
| "grad_norm": 16.55682945251465, | |
| "learning_rate": 6.363431142447469e-05, | |
| "loss": 0.233, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.4219409282700422, | |
| "grad_norm": 11.659428596496582, | |
| "learning_rate": 6.268815020619509e-05, | |
| "loss": 0.197, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.42814594192107225, | |
| "grad_norm": 20.64366912841797, | |
| "learning_rate": 6.173710723255966e-05, | |
| "loss": 0.2127, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.43435095557210224, | |
| "grad_norm": 13.794021606445312, | |
| "learning_rate": 6.078154841657027e-05, | |
| "loss": 0.2114, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.43435095557210224, | |
| "eval_avg_non_pair_similarity": 0.0004388350229078105, | |
| "eval_avg_pair_similarity": 0.010316268266644328, | |
| "eval_loss": 0.200907900929451, | |
| "eval_runtime": 52.7763, | |
| "eval_samples_per_second": 9.474, | |
| "eval_similarity_ratio": 23.508306602984025, | |
| "eval_steps_per_second": 0.303, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.4405559692231323, | |
| "grad_norm": 13.422426223754883, | |
| "learning_rate": 5.982184140869539e-05, | |
| "loss": 0.1997, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.4467609828741623, | |
| "grad_norm": 20.181900024414062, | |
| "learning_rate": 5.885835545541699e-05, | |
| "loss": 0.1893, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.45296599652519237, | |
| "grad_norm": 18.431699752807617, | |
| "learning_rate": 5.7891461257163105e-05, | |
| "loss": 0.2062, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.45917101017622236, | |
| "grad_norm": 16.194856643676758, | |
| "learning_rate": 5.6921530825681255e-05, | |
| "loss": 0.2087, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.4653760238272524, | |
| "grad_norm": 18.078935623168945, | |
| "learning_rate": 5.5948937340907156e-05, | |
| "loss": 0.1978, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.47158103747828245, | |
| "grad_norm": 20.49361801147461, | |
| "learning_rate": 5.497405500738417e-05, | |
| "loss": 0.1883, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.4777860511293125, | |
| "grad_norm": 14.588258743286133, | |
| "learning_rate": 5.399725891028824e-05, | |
| "loss": 0.1829, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.48399106478034254, | |
| "grad_norm": 14.363248825073242, | |
| "learning_rate": 5.3018924871114305e-05, | |
| "loss": 0.211, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.49019607843137253, | |
| "grad_norm": 12.266989707946777, | |
| "learning_rate": 5.203942930307929e-05, | |
| "loss": 0.1853, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.4964010920824026, | |
| "grad_norm": 13.93957805633545, | |
| "learning_rate": 5.105914906629734e-05, | |
| "loss": 0.1616, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.4964010920824026, | |
| "eval_avg_non_pair_similarity": 0.00028632804341306294, | |
| "eval_avg_pair_similarity": 0.009248366304673255, | |
| "eval_loss": 0.17835575342178345, | |
| "eval_runtime": 52.8775, | |
| "eval_samples_per_second": 9.456, | |
| "eval_similarity_ratio": 32.299896979812644, | |
| "eval_steps_per_second": 0.303, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.5026061057334326, | |
| "grad_norm": 12.441362380981445, | |
| "learning_rate": 5.007846132278327e-05, | |
| "loss": 0.1687, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.5088111193844627, | |
| "grad_norm": 17.834318161010742, | |
| "learning_rate": 4.90977433913398e-05, | |
| "loss": 0.1674, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.5150161330354927, | |
| "grad_norm": 15.299223899841309, | |
| "learning_rate": 4.8117372602384405e-05, | |
| "loss": 0.1649, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.5212211466865228, | |
| "grad_norm": 13.506607055664062, | |
| "learning_rate": 4.713772615277178e-05, | |
| "loss": 0.1833, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.5274261603375527, | |
| "grad_norm": 12.79785442352295, | |
| "learning_rate": 4.615918096066766e-05, | |
| "loss": 0.1669, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.5336311739885827, | |
| "grad_norm": 18.04769515991211, | |
| "learning_rate": 4.518211352052985e-05, | |
| "loss": 0.1611, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.5398361876396128, | |
| "grad_norm": 14.453876495361328, | |
| "learning_rate": 4.420689975825233e-05, | |
| "loss": 0.1657, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 0.5460412012906428, | |
| "grad_norm": 15.206610679626465, | |
| "learning_rate": 4.323391488652798e-05, | |
| "loss": 0.1776, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.5522462149416729, | |
| "grad_norm": 13.860987663269043, | |
| "learning_rate": 4.226353326048593e-05, | |
| "loss": 0.1491, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 0.5584512285927029, | |
| "grad_norm": 13.18816089630127, | |
| "learning_rate": 4.12961282336587e-05, | |
| "loss": 0.1426, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.5584512285927029, | |
| "eval_avg_non_pair_similarity": 0.00016228582146754844, | |
| "eval_avg_pair_similarity": 0.00711837838916108, | |
| "eval_loss": 0.16373609006404877, | |
| "eval_runtime": 52.9511, | |
| "eval_samples_per_second": 9.443, | |
| "eval_similarity_ratio": 43.86321814678376, | |
| "eval_steps_per_second": 0.302, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.5646562422437329, | |
| "grad_norm": 12.775832176208496, | |
| "learning_rate": 4.0332072014334626e-05, | |
| "loss": 0.1713, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 0.570861255894763, | |
| "grad_norm": 15.302083015441895, | |
| "learning_rate": 3.937173552235117e-05, | |
| "loss": 0.1613, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 0.577066269545793, | |
| "grad_norm": 11.7548189163208, | |
| "learning_rate": 3.841548824638376e-05, | |
| "loss": 0.1684, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 0.5832712831968231, | |
| "grad_norm": 10.636964797973633, | |
| "learning_rate": 3.7463698101785286e-05, | |
| "loss": 0.1539, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 0.5894762968478531, | |
| "grad_norm": 11.79915714263916, | |
| "learning_rate": 3.6516731289031054e-05, | |
| "loss": 0.1624, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.5956813104988831, | |
| "grad_norm": 14.726311683654785, | |
| "learning_rate": 3.557495215282342e-05, | |
| "loss": 0.1534, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 0.6018863241499132, | |
| "grad_norm": 13.101693153381348, | |
| "learning_rate": 3.463872304191036e-05, | |
| "loss": 0.1591, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 0.6080913378009432, | |
| "grad_norm": 10.77004337310791, | |
| "learning_rate": 3.3708404169672316e-05, | |
| "loss": 0.1523, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 0.6142963514519731, | |
| "grad_norm": 11.254857063293457, | |
| "learning_rate": 3.278435347553014e-05, | |
| "loss": 0.1494, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 0.6205013651030032, | |
| "grad_norm": 12.389632225036621, | |
| "learning_rate": 3.18669264872284e-05, | |
| "loss": 0.1438, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.6205013651030032, | |
| "eval_avg_non_pair_similarity": 0.00022872556135607118, | |
| "eval_avg_pair_similarity": 0.008581261232029647, | |
| "eval_loss": 0.17591074109077454, | |
| "eval_runtime": 52.6015, | |
| "eval_samples_per_second": 9.505, | |
| "eval_similarity_ratio": 37.51771853199507, | |
| "eval_steps_per_second": 0.304, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.6267063787540332, | |
| "grad_norm": 13.049078941345215, | |
| "learning_rate": 3.0956476184046275e-05, | |
| "loss": 0.1395, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 0.6329113924050633, | |
| "grad_norm": 14.119858741760254, | |
| "learning_rate": 3.0053352860989036e-05, | |
| "loss": 0.1404, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 0.6391164060560933, | |
| "grad_norm": 11.724855422973633, | |
| "learning_rate": 2.9157903994012448e-05, | |
| "loss": 0.1475, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 0.6453214197071233, | |
| "grad_norm": 16.469905853271484, | |
| "learning_rate": 2.827047410633156e-05, | |
| "loss": 0.1403, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 0.6515264333581534, | |
| "grad_norm": 11.535907745361328, | |
| "learning_rate": 2.7391404635865724e-05, | |
| "loss": 0.1324, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.6577314470091834, | |
| "grad_norm": 10.59726619720459, | |
| "learning_rate": 2.6521033803870692e-05, | |
| "loss": 0.1212, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 0.6639364606602135, | |
| "grad_norm": 20.87034034729004, | |
| "learning_rate": 2.565969648480818e-05, | |
| "loss": 0.1552, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 0.6701414743112435, | |
| "grad_norm": 9.909876823425293, | |
| "learning_rate": 2.480772407750327e-05, | |
| "loss": 0.1367, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 0.6763464879622735, | |
| "grad_norm": 13.353315353393555, | |
| "learning_rate": 2.3965444377638906e-05, | |
| "loss": 0.1447, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 0.6825515016133036, | |
| "grad_norm": 11.409924507141113, | |
| "learning_rate": 2.3133181451636858e-05, | |
| "loss": 0.1351, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.6825515016133036, | |
| "eval_avg_non_pair_similarity": 2.0496868876971767e-05, | |
| "eval_avg_pair_similarity": 0.007564989160746336, | |
| "eval_loss": 0.1368183046579361, | |
| "eval_runtime": 52.8241, | |
| "eval_samples_per_second": 9.465, | |
| "eval_similarity_ratio": 369.08023396908203, | |
| "eval_steps_per_second": 0.303, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.6887565152643336, | |
| "grad_norm": 12.820631980895996, | |
| "learning_rate": 2.2311255511973345e-05, | |
| "loss": 0.1467, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 0.6949615289153637, | |
| "grad_norm": 13.379746437072754, | |
| "learning_rate": 2.1499982793977624e-05, | |
| "loss": 0.1222, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 0.7011665425663937, | |
| "grad_norm": 12.680821418762207, | |
| "learning_rate": 2.06996754341607e-05, | |
| "loss": 0.1309, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 0.7073715562174236, | |
| "grad_norm": 12.36201286315918, | |
| "learning_rate": 1.9910641350120968e-05, | |
| "loss": 0.1231, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 0.7135765698684537, | |
| "grad_norm": 13.760623931884766, | |
| "learning_rate": 1.9133184122073305e-05, | |
| "loss": 0.122, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.7197815835194837, | |
| "grad_norm": 12.678217887878418, | |
| "learning_rate": 1.836760287604676e-05, | |
| "loss": 0.134, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 0.7259865971705137, | |
| "grad_norm": 15.280141830444336, | |
| "learning_rate": 1.761419216879601e-05, | |
| "loss": 0.1407, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 0.7321916108215438, | |
| "grad_norm": 11.139805793762207, | |
| "learning_rate": 1.687324187447102e-05, | |
| "loss": 0.1026, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 0.7383966244725738, | |
| "grad_norm": 14.420808792114258, | |
| "learning_rate": 1.6145037073088108e-05, | |
| "loss": 0.1226, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 0.7446016381236039, | |
| "grad_norm": 10.728633880615234, | |
| "learning_rate": 1.5429857940845633e-05, | |
| "loss": 0.123, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7446016381236039, | |
| "eval_avg_non_pair_similarity": 9.602570107098453e-05, | |
| "eval_avg_pair_similarity": 0.007442979822866618, | |
| "eval_loss": 0.13972054421901703, | |
| "eval_runtime": 52.699, | |
| "eval_samples_per_second": 9.488, | |
| "eval_similarity_ratio": 77.510288806583, | |
| "eval_steps_per_second": 0.304, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.7508066517746339, | |
| "grad_norm": 13.222859382629395, | |
| "learning_rate": 1.47279796423265e-05, | |
| "loss": 0.1267, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 0.7570116654256639, | |
| "grad_norm": 14.91487979888916, | |
| "learning_rate": 1.4039672224628786e-05, | |
| "loss": 0.1176, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 0.763216679076694, | |
| "grad_norm": 13.453108787536621, | |
| "learning_rate": 1.3365200513465293e-05, | |
| "loss": 0.1263, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 0.769421692727724, | |
| "grad_norm": 13.018562316894531, | |
| "learning_rate": 1.270482401127216e-05, | |
| "loss": 0.1042, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 0.7756267063787541, | |
| "grad_norm": 16.143644332885742, | |
| "learning_rate": 1.2058796797365423e-05, | |
| "loss": 0.1267, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.7818317200297841, | |
| "grad_norm": 13.684396743774414, | |
| "learning_rate": 1.1427367430184293e-05, | |
| "loss": 0.135, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 0.7880367336808141, | |
| "grad_norm": 12.124221801757812, | |
| "learning_rate": 1.0810778851658438e-05, | |
| "loss": 0.1181, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 0.7942417473318442, | |
| "grad_norm": 11.37338924407959, | |
| "learning_rate": 1.0209268293736308e-05, | |
| "loss": 0.1112, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 0.8004467609828742, | |
| "grad_norm": 13.746869087219238, | |
| "learning_rate": 9.623067187110307e-06, | |
| "loss": 0.1407, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 0.8066517746339041, | |
| "grad_norm": 13.340392112731934, | |
| "learning_rate": 9.052401072174088e-06, | |
| "loss": 0.1013, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.8066517746339041, | |
| "eval_avg_non_pair_similarity": 0.00011364468113511786, | |
| "eval_avg_pair_similarity": 0.006324736045207828, | |
| "eval_loss": 0.12133920192718506, | |
| "eval_runtime": 52.921, | |
| "eval_samples_per_second": 9.448, | |
| "eval_similarity_ratio": 55.65360368857063, | |
| "eval_steps_per_second": 0.302, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.8128567882849342, | |
| "grad_norm": 16.114587783813477, | |
| "learning_rate": 8.497489512245971e-06, | |
| "loss": 0.1231, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 0.8190618019359642, | |
| "grad_norm": 12.396312713623047, | |
| "learning_rate": 7.958546009092232e-06, | |
| "loss": 0.1194, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 0.8252668155869943, | |
| "grad_norm": 13.97695541381836, | |
| "learning_rate": 7.435777920782444e-06, | |
| "loss": 0.1055, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 0.8314718292380243, | |
| "grad_norm": 13.231639862060547, | |
| "learning_rate": 6.9293863819085865e-06, | |
| "loss": 0.1056, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 0.8376768428890543, | |
| "grad_norm": 14.493330955505371, | |
| "learning_rate": 6.43956622619874e-06, | |
| "loss": 0.1165, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.8438818565400844, | |
| "grad_norm": 14.159026145935059, | |
| "learning_rate": 5.9665059115549535e-06, | |
| "loss": 0.1158, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 0.8500868701911144, | |
| "grad_norm": 16.65438461303711, | |
| "learning_rate": 5.510387447544168e-06, | |
| "loss": 0.112, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 0.8562918838421445, | |
| "grad_norm": 10.922910690307617, | |
| "learning_rate": 5.07138632537027e-06, | |
| "loss": 0.1129, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 0.8624968974931745, | |
| "grad_norm": 11.508538246154785, | |
| "learning_rate": 4.649671450353948e-06, | |
| "loss": 0.1079, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 0.8687019111442045, | |
| "grad_norm": 13.64457893371582, | |
| "learning_rate": 4.245405076946479e-06, | |
| "loss": 0.0866, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.8687019111442045, | |
| "eval_avg_non_pair_similarity": 0.00020547029043721383, | |
| "eval_avg_pair_similarity": 0.00815738959936425, | |
| "eval_loss": 0.1222657635807991, | |
| "eval_runtime": 52.7731, | |
| "eval_samples_per_second": 9.475, | |
| "eval_similarity_ratio": 39.70106618337082, | |
| "eval_steps_per_second": 0.303, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.8749069247952346, | |
| "grad_norm": 15.069809913635254, | |
| "learning_rate": 3.858742746302535e-06, | |
| "loss": 0.1057, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 0.8811119384462646, | |
| "grad_norm": 12.91094970703125, | |
| "learning_rate": 3.489833226435818e-06, | |
| "loss": 0.1103, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 0.8873169520972947, | |
| "grad_norm": 12.701448440551758, | |
| "learning_rate": 3.138818454980641e-06, | |
| "loss": 0.1108, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 0.8935219657483247, | |
| "grad_norm": 12.398956298828125, | |
| "learning_rate": 2.8058334845816213e-06, | |
| "loss": 0.1111, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 0.8997269793993546, | |
| "grad_norm": 11.217853546142578, | |
| "learning_rate": 2.4910064309322523e-06, | |
| "loss": 0.0992, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.9059319930503847, | |
| "grad_norm": 13.966708183288574, | |
| "learning_rate": 2.194458423482465e-06, | |
| "loss": 0.1228, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 0.9121370067014147, | |
| "grad_norm": 14.061014175415039, | |
| "learning_rate": 1.9163035588342315e-06, | |
| "loss": 0.1107, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 0.9183420203524447, | |
| "grad_norm": 13.071681022644043, | |
| "learning_rate": 1.6566488568429294e-06, | |
| "loss": 0.1163, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 0.9245470340034748, | |
| "grad_norm": 12.697420120239258, | |
| "learning_rate": 1.415594219441585e-06, | |
| "loss": 0.0843, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 0.9307520476545048, | |
| "grad_norm": 10.601663589477539, | |
| "learning_rate": 1.1932323922036803e-06, | |
| "loss": 0.1068, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.9307520476545048, | |
| "eval_avg_non_pair_similarity": 0.0001585147638006401, | |
| "eval_avg_pair_similarity": 0.007807339721824974, | |
| "eval_loss": 0.11420850455760956, | |
| "eval_runtime": 52.8258, | |
| "eval_samples_per_second": 9.465, | |
| "eval_similarity_ratio": 49.25307608346225, | |
| "eval_steps_per_second": 0.303, | |
| "step": 15000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 16116, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1000, | |
| "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": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |