|
{ |
|
"best_metric": 0.513288052386553, |
|
"best_model_checkpoint": "bkai-fine-tuned-legal/checkpoint-240", |
|
"epoch": 4.0, |
|
"eval_steps": 500, |
|
"global_step": 240, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 3051630.25, |
|
"learning_rate": 2.9970400926424075e-05, |
|
"loss": 0.9663, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_dim_128_cosine_accuracy@1": 0.2080602426439827, |
|
"eval_dim_128_cosine_accuracy@10": 0.6729884256031237, |
|
"eval_dim_128_cosine_accuracy@3": 0.4049644401059824, |
|
"eval_dim_128_cosine_accuracy@5": 0.49504950495049505, |
|
"eval_dim_128_cosine_map@100": 0.34660918730461937, |
|
"eval_dim_128_cosine_mrr@10": 0.3358910891089097, |
|
"eval_dim_128_cosine_ndcg@10": 0.41500734332231626, |
|
"eval_dim_128_cosine_precision@1": 0.2080602426439827, |
|
"eval_dim_128_cosine_precision@10": 0.06729884256031238, |
|
"eval_dim_128_cosine_precision@3": 0.13498814670199413, |
|
"eval_dim_128_cosine_precision@5": 0.09900990099009901, |
|
"eval_dim_128_cosine_recall@1": 0.2080602426439827, |
|
"eval_dim_128_cosine_recall@10": 0.6729884256031237, |
|
"eval_dim_128_cosine_recall@3": 0.4049644401059824, |
|
"eval_dim_128_cosine_recall@5": 0.49504950495049505, |
|
"eval_dim_256_cosine_accuracy@1": 0.21321991354064984, |
|
"eval_dim_256_cosine_accuracy@10": 0.6862362292567284, |
|
"eval_dim_256_cosine_accuracy@3": 0.41598103472319065, |
|
"eval_dim_256_cosine_accuracy@5": 0.5063450006972529, |
|
"eval_dim_256_cosine_map@100": 0.3546268888431031, |
|
"eval_dim_256_cosine_mrr@10": 0.34420854057236694, |
|
"eval_dim_256_cosine_ndcg@10": 0.424598027833536, |
|
"eval_dim_256_cosine_precision@1": 0.21321991354064984, |
|
"eval_dim_256_cosine_precision@10": 0.06862362292567285, |
|
"eval_dim_256_cosine_precision@3": 0.1386603449077302, |
|
"eval_dim_256_cosine_precision@5": 0.10126900013945056, |
|
"eval_dim_256_cosine_recall@1": 0.21321991354064984, |
|
"eval_dim_256_cosine_recall@10": 0.6862362292567284, |
|
"eval_dim_256_cosine_recall@3": 0.41598103472319065, |
|
"eval_dim_256_cosine_recall@5": 0.5063450006972529, |
|
"eval_dim_512_cosine_accuracy@1": 0.21433551805884815, |
|
"eval_dim_512_cosine_accuracy@10": 0.6919537024124948, |
|
"eval_dim_512_cosine_accuracy@3": 0.4211407056198578, |
|
"eval_dim_512_cosine_accuracy@5": 0.508855110863199, |
|
"eval_dim_512_cosine_map@100": 0.3564257538175301, |
|
"eval_dim_512_cosine_mrr@10": 0.34621192501543907, |
|
"eval_dim_512_cosine_ndcg@10": 0.42739973393443165, |
|
"eval_dim_512_cosine_precision@1": 0.21433551805884815, |
|
"eval_dim_512_cosine_precision@10": 0.06919537024124948, |
|
"eval_dim_512_cosine_precision@3": 0.14038023520661924, |
|
"eval_dim_512_cosine_precision@5": 0.1017710221726398, |
|
"eval_dim_512_cosine_recall@1": 0.21433551805884815, |
|
"eval_dim_512_cosine_recall@10": 0.6919537024124948, |
|
"eval_dim_512_cosine_recall@3": 0.4211407056198578, |
|
"eval_dim_512_cosine_recall@5": 0.508855110863199, |
|
"eval_dim_64_cosine_accuracy@1": 0.19704364802677451, |
|
"eval_dim_64_cosine_accuracy@10": 0.6503974341096082, |
|
"eval_dim_64_cosine_accuracy@3": 0.3846046576488635, |
|
"eval_dim_64_cosine_accuracy@5": 0.48096499790824154, |
|
"eval_dim_64_cosine_map@100": 0.3332997375022508, |
|
"eval_dim_64_cosine_mrr@10": 0.3223062799237654, |
|
"eval_dim_64_cosine_ndcg@10": 0.399385228043632, |
|
"eval_dim_64_cosine_precision@1": 0.19704364802677451, |
|
"eval_dim_64_cosine_precision@10": 0.06503974341096082, |
|
"eval_dim_64_cosine_precision@3": 0.12820155254962115, |
|
"eval_dim_64_cosine_precision@5": 0.0961929995816483, |
|
"eval_dim_64_cosine_recall@1": 0.19704364802677451, |
|
"eval_dim_64_cosine_recall@10": 0.6503974341096082, |
|
"eval_dim_64_cosine_recall@3": 0.3846046576488635, |
|
"eval_dim_64_cosine_recall@5": 0.48096499790824154, |
|
"eval_dim_768_cosine_accuracy@1": 0.22116859573281272, |
|
"eval_dim_768_cosine_accuracy@10": 0.6947427137079906, |
|
"eval_dim_768_cosine_accuracy@3": 0.4196067494073351, |
|
"eval_dim_768_cosine_accuracy@5": 0.5128991772416678, |
|
"eval_dim_768_cosine_map@100": 0.3609325312259912, |
|
"eval_dim_768_cosine_mrr@10": 0.3508835189353944, |
|
"eval_dim_768_cosine_ndcg@10": 0.4315716703139609, |
|
"eval_dim_768_cosine_precision@1": 0.22116859573281272, |
|
"eval_dim_768_cosine_precision@10": 0.06947427137079905, |
|
"eval_dim_768_cosine_precision@3": 0.13986891646911168, |
|
"eval_dim_768_cosine_precision@5": 0.10257983544833356, |
|
"eval_dim_768_cosine_recall@1": 0.22116859573281272, |
|
"eval_dim_768_cosine_recall@10": 0.6947427137079906, |
|
"eval_dim_768_cosine_recall@3": 0.4196067494073351, |
|
"eval_dim_768_cosine_recall@5": 0.5128991772416678, |
|
"eval_loss": 0.30127617716789246, |
|
"eval_runtime": 4760.4689, |
|
"eval_samples_per_second": 1.507, |
|
"eval_sequential_score": 0.399385228043632, |
|
"eval_steps_per_second": 0.019, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 1582530.375, |
|
"learning_rate": 2.655769864163684e-05, |
|
"loss": 0.2399, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_dim_128_cosine_accuracy@1": 0.3225491563240831, |
|
"eval_dim_128_cosine_accuracy@10": 0.7071538139729466, |
|
"eval_dim_128_cosine_accuracy@3": 0.46102356714544696, |
|
"eval_dim_128_cosine_accuracy@5": 0.53772137777158, |
|
"eval_dim_128_cosine_map@100": 0.43211454127792864, |
|
"eval_dim_128_cosine_mrr@10": 0.42227551225947657, |
|
"eval_dim_128_cosine_ndcg@10": 0.488553211966058, |
|
"eval_dim_128_cosine_precision@1": 0.3225491563240831, |
|
"eval_dim_128_cosine_precision@10": 0.07071538139729466, |
|
"eval_dim_128_cosine_precision@3": 0.15367452238181564, |
|
"eval_dim_128_cosine_precision@5": 0.107544275554316, |
|
"eval_dim_128_cosine_recall@1": 0.3225491563240831, |
|
"eval_dim_128_cosine_recall@10": 0.7071538139729466, |
|
"eval_dim_128_cosine_recall@3": 0.46102356714544696, |
|
"eval_dim_128_cosine_recall@5": 0.53772137777158, |
|
"eval_dim_256_cosine_accuracy@1": 0.32854553060939895, |
|
"eval_dim_256_cosine_accuracy@10": 0.7131501882582625, |
|
"eval_dim_256_cosine_accuracy@3": 0.464928182959141, |
|
"eval_dim_256_cosine_accuracy@5": 0.5421837958443732, |
|
"eval_dim_256_cosine_map@100": 0.4374891918057777, |
|
"eval_dim_256_cosine_mrr@10": 0.42779830799981505, |
|
"eval_dim_256_cosine_ndcg@10": 0.49410729071963905, |
|
"eval_dim_256_cosine_precision@1": 0.32854553060939895, |
|
"eval_dim_256_cosine_precision@10": 0.07131501882582625, |
|
"eval_dim_256_cosine_precision@3": 0.1549760609863803, |
|
"eval_dim_256_cosine_precision@5": 0.10843675916887462, |
|
"eval_dim_256_cosine_recall@1": 0.32854553060939895, |
|
"eval_dim_256_cosine_recall@10": 0.7131501882582625, |
|
"eval_dim_256_cosine_recall@3": 0.464928182959141, |
|
"eval_dim_256_cosine_recall@5": 0.5421837958443732, |
|
"eval_dim_512_cosine_accuracy@1": 0.33286849811741737, |
|
"eval_dim_512_cosine_accuracy@10": 0.7142657927764607, |
|
"eval_dim_512_cosine_accuracy@3": 0.46966950216148373, |
|
"eval_dim_512_cosine_accuracy@5": 0.5445544554455446, |
|
"eval_dim_512_cosine_map@100": 0.4410220731771345, |
|
"eval_dim_512_cosine_mrr@10": 0.4314777000838923, |
|
"eval_dim_512_cosine_ndcg@10": 0.4972638536904392, |
|
"eval_dim_512_cosine_precision@1": 0.33286849811741737, |
|
"eval_dim_512_cosine_precision@10": 0.07142657927764608, |
|
"eval_dim_512_cosine_precision@3": 0.15655650072049457, |
|
"eval_dim_512_cosine_precision@5": 0.10891089108910891, |
|
"eval_dim_512_cosine_recall@1": 0.33286849811741737, |
|
"eval_dim_512_cosine_recall@10": 0.7142657927764607, |
|
"eval_dim_512_cosine_recall@3": 0.46966950216148373, |
|
"eval_dim_512_cosine_recall@5": 0.5445544554455446, |
|
"eval_dim_64_cosine_accuracy@1": 0.306233440245433, |
|
"eval_dim_64_cosine_accuracy@10": 0.6906986473295217, |
|
"eval_dim_64_cosine_accuracy@3": 0.44205829033607585, |
|
"eval_dim_64_cosine_accuracy@5": 0.5208478594338307, |
|
"eval_dim_64_cosine_map@100": 0.416048690892794, |
|
"eval_dim_64_cosine_mrr@10": 0.4057003738603239, |
|
"eval_dim_64_cosine_ndcg@10": 0.47195343698452114, |
|
"eval_dim_64_cosine_precision@1": 0.306233440245433, |
|
"eval_dim_64_cosine_precision@10": 0.06906986473295217, |
|
"eval_dim_64_cosine_precision@3": 0.1473527634453586, |
|
"eval_dim_64_cosine_precision@5": 0.10416957188676614, |
|
"eval_dim_64_cosine_recall@1": 0.306233440245433, |
|
"eval_dim_64_cosine_recall@10": 0.6906986473295217, |
|
"eval_dim_64_cosine_recall@3": 0.44205829033607585, |
|
"eval_dim_64_cosine_recall@5": 0.5208478594338307, |
|
"eval_dim_768_cosine_accuracy@1": 0.33412355320039044, |
|
"eval_dim_768_cosine_accuracy@10": 0.7152419467298843, |
|
"eval_dim_768_cosine_accuracy@3": 0.47106400780923163, |
|
"eval_dim_768_cosine_accuracy@5": 0.5448333565750941, |
|
"eval_dim_768_cosine_map@100": 0.44207831994901653, |
|
"eval_dim_768_cosine_mrr@10": 0.4325288363846451, |
|
"eval_dim_768_cosine_ndcg@10": 0.49828039957821635, |
|
"eval_dim_768_cosine_precision@1": 0.33412355320039044, |
|
"eval_dim_768_cosine_precision@10": 0.07152419467298843, |
|
"eval_dim_768_cosine_precision@3": 0.15702133593641054, |
|
"eval_dim_768_cosine_precision@5": 0.10896667131501885, |
|
"eval_dim_768_cosine_recall@1": 0.33412355320039044, |
|
"eval_dim_768_cosine_recall@10": 0.7152419467298843, |
|
"eval_dim_768_cosine_recall@3": 0.47106400780923163, |
|
"eval_dim_768_cosine_recall@5": 0.5448333565750941, |
|
"eval_loss": 0.18898583948612213, |
|
"eval_runtime": 4757.4, |
|
"eval_samples_per_second": 1.508, |
|
"eval_sequential_score": 0.47195343698452114, |
|
"eval_steps_per_second": 0.019, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"grad_norm": 792357.3125, |
|
"learning_rate": 1.8730348307472828e-05, |
|
"loss": 0.1032, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_dim_128_cosine_accuracy@1": 0.3468135545948961, |
|
"eval_dim_128_cosine_accuracy@10": 0.7173337052015061, |
|
"eval_dim_128_cosine_accuracy@3": 0.46520708408869055, |
|
"eval_dim_128_cosine_accuracy@5": 0.5407892901966252, |
|
"eval_dim_128_cosine_map@100": 0.447677926706174, |
|
"eval_dim_128_cosine_mrr@10": 0.43798207285517343, |
|
"eval_dim_128_cosine_ndcg@10": 0.5026334575249692, |
|
"eval_dim_128_cosine_precision@1": 0.3468135545948961, |
|
"eval_dim_128_cosine_precision@10": 0.07173337052015061, |
|
"eval_dim_128_cosine_precision@3": 0.15506902802956352, |
|
"eval_dim_128_cosine_precision@5": 0.10815785803932505, |
|
"eval_dim_128_cosine_recall@1": 0.3468135545948961, |
|
"eval_dim_128_cosine_recall@10": 0.7173337052015061, |
|
"eval_dim_128_cosine_recall@3": 0.46520708408869055, |
|
"eval_dim_128_cosine_recall@5": 0.5407892901966252, |
|
"eval_dim_256_cosine_accuracy@1": 0.35043926927904057, |
|
"eval_dim_256_cosine_accuracy@10": 0.7156602984242086, |
|
"eval_dim_256_cosine_accuracy@3": 0.4700878538558081, |
|
"eval_dim_256_cosine_accuracy@5": 0.544972807139869, |
|
"eval_dim_256_cosine_map@100": 0.4507378853512356, |
|
"eval_dim_256_cosine_mrr@10": 0.44098468699988747, |
|
"eval_dim_256_cosine_ndcg@10": 0.5046872072161374, |
|
"eval_dim_256_cosine_precision@1": 0.35043926927904057, |
|
"eval_dim_256_cosine_precision@10": 0.07156602984242087, |
|
"eval_dim_256_cosine_precision@3": 0.15669595128526936, |
|
"eval_dim_256_cosine_precision@5": 0.1089945614279738, |
|
"eval_dim_256_cosine_recall@1": 0.35043926927904057, |
|
"eval_dim_256_cosine_recall@10": 0.7156602984242086, |
|
"eval_dim_256_cosine_recall@3": 0.4700878538558081, |
|
"eval_dim_256_cosine_recall@5": 0.544972807139869, |
|
"eval_dim_512_cosine_accuracy@1": 0.35169432436201364, |
|
"eval_dim_512_cosine_accuracy@10": 0.7231906289220471, |
|
"eval_dim_512_cosine_accuracy@3": 0.4700878538558081, |
|
"eval_dim_512_cosine_accuracy@5": 0.5448333565750941, |
|
"eval_dim_512_cosine_map@100": 0.45212818188900633, |
|
"eval_dim_512_cosine_mrr@10": 0.44279272997722396, |
|
"eval_dim_512_cosine_ndcg@10": 0.5076751608428753, |
|
"eval_dim_512_cosine_precision@1": 0.35169432436201364, |
|
"eval_dim_512_cosine_precision@10": 0.07231906289220472, |
|
"eval_dim_512_cosine_precision@3": 0.15669595128526936, |
|
"eval_dim_512_cosine_precision@5": 0.10896667131501885, |
|
"eval_dim_512_cosine_recall@1": 0.35169432436201364, |
|
"eval_dim_512_cosine_recall@10": 0.7231906289220471, |
|
"eval_dim_512_cosine_recall@3": 0.4700878538558081, |
|
"eval_dim_512_cosine_recall@5": 0.5448333565750941, |
|
"eval_dim_64_cosine_accuracy@1": 0.3356575094129131, |
|
"eval_dim_64_cosine_accuracy@10": 0.6976711755682611, |
|
"eval_dim_64_cosine_accuracy@3": 0.4523776321294101, |
|
"eval_dim_64_cosine_accuracy@5": 0.5307488495328406, |
|
"eval_dim_64_cosine_map@100": 0.43559811808588345, |
|
"eval_dim_64_cosine_mrr@10": 0.42538852476797007, |
|
"eval_dim_64_cosine_ndcg@10": 0.48852868771634256, |
|
"eval_dim_64_cosine_precision@1": 0.3356575094129131, |
|
"eval_dim_64_cosine_precision@10": 0.0697671175568261, |
|
"eval_dim_64_cosine_precision@3": 0.1507925440431367, |
|
"eval_dim_64_cosine_precision@5": 0.10614976990656813, |
|
"eval_dim_64_cosine_recall@1": 0.3356575094129131, |
|
"eval_dim_64_cosine_recall@10": 0.6976711755682611, |
|
"eval_dim_64_cosine_recall@3": 0.4523776321294101, |
|
"eval_dim_64_cosine_recall@5": 0.5307488495328406, |
|
"eval_dim_768_cosine_accuracy@1": 0.3568539952586808, |
|
"eval_dim_768_cosine_accuracy@10": 0.7237484311811463, |
|
"eval_dim_768_cosine_accuracy@3": 0.473434667410403, |
|
"eval_dim_768_cosine_accuracy@5": 0.5530609398968066, |
|
"eval_dim_768_cosine_map@100": 0.456489701012935, |
|
"eval_dim_768_cosine_mrr@10": 0.44698465822437466, |
|
"eval_dim_768_cosine_ndcg@10": 0.5110884733352402, |
|
"eval_dim_768_cosine_precision@1": 0.3568539952586808, |
|
"eval_dim_768_cosine_precision@10": 0.07237484311811462, |
|
"eval_dim_768_cosine_precision@3": 0.15781155580346765, |
|
"eval_dim_768_cosine_precision@5": 0.11061218797936133, |
|
"eval_dim_768_cosine_recall@1": 0.3568539952586808, |
|
"eval_dim_768_cosine_recall@10": 0.7237484311811463, |
|
"eval_dim_768_cosine_recall@3": 0.473434667410403, |
|
"eval_dim_768_cosine_recall@5": 0.5530609398968066, |
|
"eval_loss": 0.15947215259075165, |
|
"eval_runtime": 4772.3881, |
|
"eval_samples_per_second": 1.503, |
|
"eval_sequential_score": 0.48852868771634256, |
|
"eval_steps_per_second": 0.019, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 528732.6875, |
|
"learning_rate": 9.478131709729831e-06, |
|
"loss": 0.0562, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_dim_128_cosine_accuracy@1": 0.35016036814949103, |
|
"eval_dim_128_cosine_accuracy@10": 0.71803095802538, |
|
"eval_dim_128_cosine_accuracy@3": 0.466880490865988, |
|
"eval_dim_128_cosine_accuracy@5": 0.5453911588341933, |
|
"eval_dim_128_cosine_map@100": 0.45032612042753656, |
|
"eval_dim_128_cosine_mrr@10": 0.44050070500007754, |
|
"eval_dim_128_cosine_ndcg@10": 0.5047593926261945, |
|
"eval_dim_128_cosine_precision@1": 0.35016036814949103, |
|
"eval_dim_128_cosine_precision@10": 0.07180309580253802, |
|
"eval_dim_128_cosine_precision@3": 0.15562683028866267, |
|
"eval_dim_128_cosine_precision@5": 0.10907823176683866, |
|
"eval_dim_128_cosine_recall@1": 0.35016036814949103, |
|
"eval_dim_128_cosine_recall@10": 0.71803095802538, |
|
"eval_dim_128_cosine_recall@3": 0.466880490865988, |
|
"eval_dim_128_cosine_recall@5": 0.5453911588341933, |
|
"eval_dim_256_cosine_accuracy@1": 0.34932366476084226, |
|
"eval_dim_256_cosine_accuracy@10": 0.7227722772277227, |
|
"eval_dim_256_cosine_accuracy@3": 0.46660158973643845, |
|
"eval_dim_256_cosine_accuracy@5": 0.5445544554455446, |
|
"eval_dim_256_cosine_map@100": 0.4503116262429447, |
|
"eval_dim_256_cosine_mrr@10": 0.44069499505282583, |
|
"eval_dim_256_cosine_ndcg@10": 0.5059436181895386, |
|
"eval_dim_256_cosine_precision@1": 0.34932366476084226, |
|
"eval_dim_256_cosine_precision@10": 0.07227722772277227, |
|
"eval_dim_256_cosine_precision@3": 0.15553386324547946, |
|
"eval_dim_256_cosine_precision@5": 0.10891089108910891, |
|
"eval_dim_256_cosine_recall@1": 0.34932366476084226, |
|
"eval_dim_256_cosine_recall@10": 0.7227722772277227, |
|
"eval_dim_256_cosine_recall@3": 0.46660158973643845, |
|
"eval_dim_256_cosine_recall@5": 0.5445544554455446, |
|
"eval_dim_512_cosine_accuracy@1": 0.3550411379166086, |
|
"eval_dim_512_cosine_accuracy@10": 0.7234695300515968, |
|
"eval_dim_512_cosine_accuracy@3": 0.4693906010319342, |
|
"eval_dim_512_cosine_accuracy@5": 0.5484590712592387, |
|
"eval_dim_512_cosine_map@100": 0.45428241375668854, |
|
"eval_dim_512_cosine_mrr@10": 0.4447460339595335, |
|
"eval_dim_512_cosine_ndcg@10": 0.5091968606671677, |
|
"eval_dim_512_cosine_precision@1": 0.3550411379166086, |
|
"eval_dim_512_cosine_precision@10": 0.07234695300515968, |
|
"eval_dim_512_cosine_precision@3": 0.1564635336773114, |
|
"eval_dim_512_cosine_precision@5": 0.10969181425184772, |
|
"eval_dim_512_cosine_recall@1": 0.3550411379166086, |
|
"eval_dim_512_cosine_recall@10": 0.7234695300515968, |
|
"eval_dim_512_cosine_recall@3": 0.4693906010319342, |
|
"eval_dim_512_cosine_recall@5": 0.5484590712592387, |
|
"eval_dim_64_cosine_accuracy@1": 0.33760981731976014, |
|
"eval_dim_64_cosine_accuracy@10": 0.7082694184911449, |
|
"eval_dim_64_cosine_accuracy@3": 0.45349323664760843, |
|
"eval_dim_64_cosine_accuracy@5": 0.5308883000976153, |
|
"eval_dim_64_cosine_map@100": 0.4383018870756752, |
|
"eval_dim_64_cosine_mrr@10": 0.4285745828103948, |
|
"eval_dim_64_cosine_ndcg@10": 0.49330694449266793, |
|
"eval_dim_64_cosine_precision@1": 0.33760981731976014, |
|
"eval_dim_64_cosine_precision@10": 0.07082694184911449, |
|
"eval_dim_64_cosine_precision@3": 0.1511644122158695, |
|
"eval_dim_64_cosine_precision@5": 0.10617766001952307, |
|
"eval_dim_64_cosine_recall@1": 0.33760981731976014, |
|
"eval_dim_64_cosine_recall@10": 0.7082694184911449, |
|
"eval_dim_64_cosine_recall@3": 0.45349323664760843, |
|
"eval_dim_64_cosine_recall@5": 0.5308883000976153, |
|
"eval_dim_768_cosine_accuracy@1": 0.36061916050760007, |
|
"eval_dim_768_cosine_accuracy@10": 0.7238878817459211, |
|
"eval_dim_768_cosine_accuracy@3": 0.478036536047971, |
|
"eval_dim_768_cosine_accuracy@5": 0.5529214893320318, |
|
"eval_dim_768_cosine_map@100": 0.45954079808905995, |
|
"eval_dim_768_cosine_mrr@10": 0.44983946583793183, |
|
"eval_dim_768_cosine_ndcg@10": 0.513288052386553, |
|
"eval_dim_768_cosine_precision@1": 0.36061916050760007, |
|
"eval_dim_768_cosine_precision@10": 0.07238878817459211, |
|
"eval_dim_768_cosine_precision@3": 0.15934551201599031, |
|
"eval_dim_768_cosine_precision@5": 0.11058429786640636, |
|
"eval_dim_768_cosine_recall@1": 0.36061916050760007, |
|
"eval_dim_768_cosine_recall@10": 0.7238878817459211, |
|
"eval_dim_768_cosine_recall@3": 0.478036536047971, |
|
"eval_dim_768_cosine_recall@5": 0.5529214893320318, |
|
"eval_loss": 0.14603222906589508, |
|
"eval_runtime": 4765.1388, |
|
"eval_samples_per_second": 1.505, |
|
"eval_sequential_score": 0.49330694449266793, |
|
"eval_steps_per_second": 0.019, |
|
"step": 240 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 354, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 6, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"EarlyStoppingCallback": { |
|
"args": { |
|
"early_stopping_patience": 1, |
|
"early_stopping_threshold": 0.001 |
|
}, |
|
"attributes": { |
|
"early_stopping_patience_counter": 0 |
|
} |
|
}, |
|
"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": 80, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|