Training in progress, step 5000

config.json

@@ -4,7 +4,7 @@
   ],
   "attention_probs_dropout_prob": 0.1,
   "classifier_dropout": null,
-  "dtype": "float32",
+  "dtype": "bfloat16",
   "gradient_checkpointing": false,
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,

eval/Information-Retrieval_evaluation_test_results.csv

@@ -164,3 +164,34 @@ epoch,steps,cosine-Accuracy@1,cosine-Precision@1,cosine-Recall@1,cosine-MRR@1,co
 0.069203087995571,9000,0.5392995765702188,0.5392995765702188,0.5205514668817421,0.5392995765702188,0.7411493116676775,0.6842907825191141,{1: np.float64(0.3076407769656779)},{1: np.float64(0.1617637908660203)}
 0.07304770399532495,9500,0.5377999294283698,0.5377999294283698,0.519077181419498,0.5377999294283698,0.740036819463065,0.6829700307709706,{1: np.float64(0.30738499557813687)},{1: np.float64(0.16471705861501704)}
 0.0768923199950789,10000,0.5365649258997883,0.5365649258997883,0.5179219385354706,0.5365649258997883,0.73967794014998,0.6824233005982829,{1: np.float64(0.3053861957229551)},{1: np.float64(0.1622466592814671)}
+0.08073693599483284,10500,0.5377999294283698,0.5377999294283698,0.5190595385119467,0.5377999294283698,0.740209454449655,0.6830983582120237,{1: np.float64(0.3076015298360008)},{1: np.float64(0.16244411574097498)}
+0.08458155199458678,11000,0.5371383203952012,0.5371383203952012,0.5183924160701684,0.5371383203952012,0.7395353004528856,0.6823944894496982,{1: np.float64(0.30709089767387115)},{1: np.float64(0.1627062390303891)}
+0.08842616799434072,11500,0.5358592095977417,0.5358592095977417,0.5170544955808717,0.5358592095977417,0.73905673108783,0.6816042576476347,{1: np.float64(0.3056584182224808)},{1: np.float64(0.1635555963756791)}
+0.09227078399409468,12000,0.5343154551870148,0.5343154551870148,0.5154901577780019,0.5343154551870148,0.7380762591404793,0.6804053205694645,{1: np.float64(0.303617687098401)},{1: np.float64(0.1686003703545312)}
+0,0,0.5474153140437544,0.5474153140437544,0.5284661487213093,0.5474153140437544,0.7464497171316696,0.6905122350847523,{1: np.float64(0.31534550033472736)},{1: np.float64(0.16084113241756684)}
+0.0038446159997539446,500,0.5474153140437544,0.5474153140437544,0.5284661487213093,0.5474153140437544,0.7464720677396254,0.6905619587330443,{1: np.float64(0.315738610374166)},{1: np.float64(0.16085064599971574)}
+0.007689231999507889,1000,0.5473270995059986,0.5473270995059986,0.5283779341835534,0.5473270995059986,0.7464247700543576,0.6904797504901512,{1: np.float64(0.3153540061284735)},{1: np.float64(0.16065309629567)}
+0.011533847999261835,1500,0.5475035285815102,0.5475035285815102,0.5285617144705448,0.5475035285815102,0.7465334090898501,0.6906071372941966,{1: np.float64(0.3156745850146082)},{1: np.float64(0.16072006910244244)}
+0,0,0.5474394601032155,0.5474394601032155,0.5284894589479743,0.5474394601032155,0.7464232866184599,0.6905199963377163,{1: np.float64(0.31524254043885996)},{1: np.float64(0.16089488030492544)}
+0,0,0.5474394601032155,0.5474394601032155,0.5284894589479743,0.5474394601032155,0.7464232866184599,0.6905199963377163,{1: np.float64(0.31524254043885996)},{1: np.float64(0.16089488030492544)}
+0,0,0.5474394601032155,0.5474394601032155,0.5284894589479743,0.5474394601032155,0.7464232866184599,0.6905199963377163,{1: np.float64(0.31524254043885996)},{1: np.float64(0.16089488030492544)}
+0.01711039627677777,500,0.5474394601032155,0.5474394601032155,0.5285041620194458,0.5474394601032155,0.7464750438086984,0.6905674076498488,{1: np.float64(0.3148968570933055)},{1: np.float64(0.16082601620043113)}
+0.03422079255355554,1000,0.5477482246041198,0.5477482246041198,0.5287982234488785,0.5477482246041198,0.7466339876248261,0.6907467085004655,{1: np.float64(0.31574441364982964)},{1: np.float64(0.16051035580843753)}
+0.05133118883033331,1500,0.548498081249173,0.548498081249173,0.5295554316296675,0.548498081249173,0.7468970269854199,0.6911438656835909,{1: np.float64(0.3163926415330103)},{1: np.float64(0.16023465676374488)}
+0.06844158510711108,2000,0.5492038286798112,0.5492038286798112,0.5303699817891957,0.5492038286798112,0.7472779153918623,0.69166420585171,{1: np.float64(0.31771636053828783)},{1: np.float64(0.1601896817287953)}
+0.08555198138388885,2500,0.5506594327555027,0.5506594327555027,0.531832937400623,0.5506594327555027,0.7478255402442747,0.6924532879194532,{1: np.float64(0.31955962037044483)},{1: np.float64(0.16052405988742702)}
+0,0,0.5474394601032155,0.5474394601032155,0.5285041620194458,0.5474394601032155,0.746427629996024,0.690521727376417,{1: np.float64(0.3146675138109062)},{1: np.float64(0.16090423024805375)}
+0.0684369011771147,500,0.54766000617529,0.54766000617529,0.5286438411984263,0.54766000617529,0.7465719977250953,0.6906841922591135,{1: np.float64(0.315027988525578)},{1: np.float64(0.16086781810556405)}
+0.1368738023542294,1000,0.5478364430329495,0.5478364430329495,0.5288937934134441,0.5478364430329495,0.7466781914555873,0.6908147857710708,{1: np.float64(0.3152265565629674)},{1: np.float64(0.16043224497901973)}
+0,0,0.5833362654375286,0.5833362654375286,0.565457315116458,0.5833362654375286,0.7700534613587743,0.7186778814620468,{1: np.float64(0.3409873464629461)},{1: np.float64(0.15002254468648485)}
+0.04066197698532103,500,0.5831955244361564,0.5831955244361564,0.5652696604479617,0.5831955244361564,0.7699919719772824,0.7186269552794476,{1: np.float64(0.3407495163383317)},{1: np.float64(0.15018024680933917)}
+0.08132395397064206,1000,0.5834066359382147,0.5834066359382147,0.5655015222258633,0.5834066359382147,0.7702120717451748,0.7187244367996998,{1: np.float64(0.34094261270103304)},{1: np.float64(0.1508126746150439)}
+0.12198593095596308,1500,0.5832658949368424,0.5832658949368424,0.5653610744349107,0.5832658949368424,0.7700855775813362,0.7187280208527718,{1: np.float64(0.3415084448173125)},{1: np.float64(0.15191957801939823)}
+0.16264790794128411,2000,0.5828084866823827,0.5828084866823827,0.5649494070058969,0.5828084866823827,0.7696093167773582,0.718331496556978,{1: np.float64(0.34123791177319884)},{1: np.float64(0.15265015868808565)}
+0,0,0.5738386045345464,0.5738386045345464,0.5557289188817002,0.5738386045345464,0.7643034040823512,0.7118104615259053,{1: None},{1: None}
+0,0,0.5738386045345464,0.5738386045345464,0.5557289188817002,0.5738386045345464,0.7643034040823512,0.7118104615259053
+0.02486139770777913,1000,0.5739460582399083,0.5739460582399083,0.5558512967128068,0.5739460582399083,0.7643523165048973,0.7118479091075285
+0.04972279541555826,2000,0.5747698699810165,0.5747698699810165,0.556620417273109,0.5747698699810165,0.7643733580225612,0.712045315319712
+0.07458419312333739,3000,0.5748773236863784,0.5748773236863784,0.5569231474237922,0.5748773236863784,0.7633121170569411,0.7115701116310537
+0.09944559083111652,4000,0.5727998853827143,0.5727998853827143,0.5551358341246052,0.5727998853827143,0.7597610382174612,0.7085525073986481
+0.12430698853889566,5000,0.5736236971238224,0.5736236971238224,0.5561059022980116,0.5736236971238224,0.758749779559982,0.7079949743990421

final_metrics.json

@@ -1,10 +1,10 @@
 {
-    "test_cosine_accuracy@1": 0.5488267466478476,
-    "test_cosine_precision@1": 0.5488267466478476,
-    "test_cosine_recall@1": 0.5299124995799307,
-    "test_cosine_ndcg@10": 0.7468543359493904,
-    "test_cosine_mrr@1": 0.5488267466478476,
-    "test_cosine_map@100": 0.6913077014908864,
-    "test_cosine_auc_precision_cache_hit_ratio": 0.3205152119302144,
-    "test_cosine_auc_similarity_distribution": 0.15285602105071638
+    "test_cosine_accuracy@1": 0.5474394601032155,
+    "test_cosine_precision@1": 0.5474394601032155,
+    "test_cosine_recall@1": 0.5284894589479743,
+    "test_cosine_ndcg@10": 0.7464605932142311,
+    "test_cosine_mrr@1": 0.5474394601032155,
+    "test_cosine_map@100": 0.6905991135837425,
+    "test_cosine_auc_precision_cache_hit_ratio": 0.3154094226830552,
+    "test_cosine_auc_similarity_distribution": 0.16081146552358963
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9256db3f3e9170f5e60d958aa67da5f2a6a71e45a24165c8dd916f78af687726
-size 90864192
+oid sha256:039464dd75005fb8a767e9c573b8b125b258ea4decf8c2c986f70be531f671b0
+size 45437864

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:25903747f0c9677eda621170598a7ec5a21945e67d634603f9c55b7b57679f18
+oid sha256:891aa5ca12846256a58ad7de8c47eb42258f9c14f98b7a5bce97377f455f05d7
 size 6353