Hugging Face's logo

aisuko
/
training

Sentence Similarity
sentence-transformers
TensorBoard
Safetensors
bert
feature-extraction
Generated from Trainer
dataset_size:4984
loss:MultipleNegativesRankingLoss
Eval Results
text-embeddings-inference
Model card Files
xet
Metrics Community
aisuko commited on
Commit
573158c
·
verified ·
1 Parent(s): 66c3c7b

aisuko/encoder-L6-V2

Browse files
Files changed (5) hide show
  1. README.md +62 -17
  2. eval/similarity_evaluation_sts-dev_results.csv +50 -5
  3. model.safetensors +1 -1
  4. runs/Jul30_13-09-19_4e06a36bb887/events.out.tfevents.1753880978.4e06a36bb887.13.0 +3 -0
  5. training_args.bin +1 -1
README.md CHANGED
@@ -371,10 +371,10 @@ model-index:
371
  type: sts-dev
372
  metrics:
373
  - type: pearson_cosine
374
- value: 0.8776143775032945
375
  name: Pearson Cosine
376
  - type: spearman_cosine
377
- value: 0.8754887541943006
378
  name: Spearman Cosine
379
  - task:
380
  type: semantic-similarity
@@ -384,10 +384,10 @@ model-index:
384
  type: sts-test
385
  metrics:
386
  - type: pearson_cosine
387
- value: 0.877613066248591
388
  name: Pearson Cosine
389
  - type: spearman_cosine
390
- value: 0.875487017524482
391
  name: Spearman Cosine
392
  ---
393
 
@@ -491,8 +491,8 @@ You can finetune this model on your own dataset.
491
 
492
  | Metric | sts-dev | sts-test |
493
  |:--------------------|:-----------|:-----------|
494
- | pearson_cosine | 0.8776 | 0.8776 |
495
- | **spearman_cosine** | **0.8755** | **0.8755** |
496
 
497
  <!--
498
  ## Bias, Risks and Limitations
@@ -567,7 +567,7 @@ You can finetune this model on your own dataset.
567
  - `per_device_train_batch_size`: 16
568
  - `per_device_eval_batch_size`: 16
569
  - `learning_rate`: 2e-05
570
- - `num_train_epochs`: 2
571
  - `warmup_ratio`: 0.1
572
  - `fp16`: True
573
  - `batch_sampler`: no_duplicates
@@ -592,7 +592,7 @@ You can finetune this model on your own dataset.
592
  - `adam_beta2`: 0.999
593
  - `adam_epsilon`: 1e-08
594
  - `max_grad_norm`: 1.0
595
- - `num_train_epochs`: 2
596
  - `max_steps`: -1
597
  - `lr_scheduler_type`: linear
598
  - `lr_scheduler_kwargs`: {}
@@ -693,15 +693,60 @@ You can finetune this model on your own dataset.
693
  </details>
694
 
695
  ### Training Logs
696
- | Epoch | Step | Training Loss | Validation Loss | sts-dev_spearman_cosine | sts-test_spearman_cosine |
697
- |:------:|:----:|:-------------:|:---------------:|:-----------------------:|:------------------------:|
698
- | -1 | -1 | - | - | 0.8671 | - |
699
- | 0.3953 | 100 | 0.0204 | 0.0018 | 0.8743 | - |
700
- | 0.7905 | 200 | 0.0046 | 0.0020 | 0.8752 | - |
701
- | 1.1858 | 300 | 0.0021 | 0.0012 | 0.8752 | - |
702
- | 1.5810 | 400 | 0.0008 | 0.0016 | 0.8757 | - |
703
- | 1.9763 | 500 | 0.0027 | 0.0014 | 0.8755 | - |
704
- | -1 | -1 | - | - | - | 0.8755 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
705
 
706
 
707
  ### Framework Versions
 
371
  type: sts-dev
372
  metrics:
373
  - type: pearson_cosine
374
+ value: 0.8793284892973376
375
  name: Pearson Cosine
376
  - type: spearman_cosine
377
+ value: 0.876484495899188
378
  name: Spearman Cosine
379
  - task:
380
  type: semantic-similarity
 
384
  type: sts-test
385
  metrics:
386
  - type: pearson_cosine
387
+ value: 0.879334132854901
388
  name: Pearson Cosine
389
  - type: spearman_cosine
390
+ value: 0.8764936381058213
391
  name: Spearman Cosine
392
  ---
393
 
 
491
 
492
  | Metric | sts-dev | sts-test |
493
  |:--------------------|:-----------|:-----------|
494
+ | pearson_cosine | 0.8793 | 0.8793 |
495
+ | **spearman_cosine** | **0.8765** | **0.8765** |
496
 
497
  <!--
498
  ## Bias, Risks and Limitations
 
567
  - `per_device_train_batch_size`: 16
568
  - `per_device_eval_batch_size`: 16
569
  - `learning_rate`: 2e-05
570
+ - `num_train_epochs`: 20
571
  - `warmup_ratio`: 0.1
572
  - `fp16`: True
573
  - `batch_sampler`: no_duplicates
 
592
  - `adam_beta2`: 0.999
593
  - `adam_epsilon`: 1e-08
594
  - `max_grad_norm`: 1.0
595
+ - `num_train_epochs`: 20
596
  - `max_steps`: -1
597
  - `lr_scheduler_type`: linear
598
  - `lr_scheduler_kwargs`: {}
 
693
  </details>
694
 
695
  ### Training Logs
696
+ | Epoch | Step | Training Loss | Validation Loss | sts-dev_spearman_cosine | sts-test_spearman_cosine |
697
+ |:-------:|:----:|:-------------:|:---------------:|:-----------------------:|:------------------------:|
698
+ | -1 | -1 | - | - | 0.8671 | - |
699
+ | 0.3953 | 100 | 0.0422 | 0.0031 | 0.8701 | - |
700
+ | 0.7905 | 200 | 0.0105 | 0.0017 | 0.8727 | - |
701
+ | 1.1858 | 300 | 0.0041 | 0.0016 | 0.8728 | - |
702
+ | 1.5810 | 400 | 0.0016 | 0.0011 | 0.8730 | - |
703
+ | 1.9763 | 500 | 0.0039 | 0.0021 | 0.8731 | - |
704
+ | 2.3715 | 600 | 0.0014 | 0.0020 | 0.8741 | - |
705
+ | 2.7668 | 700 | 0.0014 | 0.0017 | 0.8744 | - |
706
+ | 3.1621 | 800 | 0.0019 | 0.0009 | 0.8742 | - |
707
+ | 3.5573 | 900 | 0.0012 | 0.0011 | 0.8754 | - |
708
+ | 3.9526 | 1000 | 0.0016 | 0.0015 | 0.8760 | - |
709
+ | 4.3478 | 1100 | 0.0021 | 0.0011 | 0.8763 | - |
710
+ | 4.7431 | 1200 | 0.0006 | 0.0009 | 0.8753 | - |
711
+ | 5.1383 | 1300 | 0.0004 | 0.0009 | 0.8753 | - |
712
+ | 5.5336 | 1400 | 0.0008 | 0.0008 | 0.8751 | - |
713
+ | 5.9289 | 1500 | 0.0004 | 0.0004 | 0.8743 | - |
714
+ | 6.3241 | 1600 | 0.0009 | 0.0008 | 0.8758 | - |
715
+ | 6.7194 | 1700 | 0.0005 | 0.0009 | 0.8747 | - |
716
+ | 7.1146 | 1800 | 0.0004 | 0.0006 | 0.8742 | - |
717
+ | 7.5099 | 1900 | 0.0003 | 0.0010 | 0.8748 | - |
718
+ | 7.9051 | 2000 | 0.0006 | 0.0008 | 0.8742 | - |
719
+ | 8.3004 | 2100 | 0.0005 | 0.0007 | 0.8744 | - |
720
+ | 8.6957 | 2200 | 0.0003 | 0.0006 | 0.8748 | - |
721
+ | 9.0909 | 2300 | 0.0005 | 0.0012 | 0.8749 | - |
722
+ | 9.4862 | 2400 | 0.0007 | 0.0006 | 0.8762 | - |
723
+ | 9.8814 | 2500 | 0.0003 | 0.0009 | 0.8762 | - |
724
+ | 10.2767 | 2600 | 0.0004 | 0.0007 | 0.8759 | - |
725
+ | 10.6719 | 2700 | 0.0005 | 0.0005 | 0.8760 | - |
726
+ | 11.0672 | 2800 | 0.0005 | 0.0007 | 0.8754 | - |
727
+ | 11.4625 | 2900 | 0.0002 | 0.0008 | 0.8749 | - |
728
+ | 11.8577 | 3000 | 0.0002 | 0.0007 | 0.8749 | - |
729
+ | 12.2530 | 3100 | 0.0003 | 0.0007 | 0.8752 | - |
730
+ | 12.6482 | 3200 | 0.0004 | 0.0008 | 0.8760 | - |
731
+ | 13.0435 | 3300 | 0.0002 | 0.0008 | 0.8767 | - |
732
+ | 13.4387 | 3400 | 0.0002 | 0.0007 | 0.8763 | - |
733
+ | 13.8340 | 3500 | 0.0002 | 0.0007 | 0.8763 | - |
734
+ | 14.2292 | 3600 | 0.0001 | 0.0007 | 0.8764 | - |
735
+ | 14.6245 | 3700 | 0.0003 | 0.0006 | 0.8765 | - |
736
+ | 15.0198 | 3800 | 0.0002 | 0.0005 | 0.8757 | - |
737
+ | 15.4150 | 3900 | 0.0002 | 0.0004 | 0.8760 | - |
738
+ | 15.8103 | 4000 | 0.0002 | 0.0005 | 0.8765 | - |
739
+ | 16.2055 | 4100 | 0.0002 | 0.0005 | 0.8757 | - |
740
+ | 16.6008 | 4200 | 0.0002 | 0.0006 | 0.8758 | - |
741
+ | 16.9960 | 4300 | 0.0002 | 0.0006 | 0.8758 | - |
742
+ | 17.3913 | 4400 | 0.0001 | 0.0005 | 0.8761 | - |
743
+ | 17.7866 | 4500 | 0.0002 | 0.0005 | 0.8765 | - |
744
+ | 18.1818 | 4600 | 0.0001 | 0.0005 | 0.8767 | - |
745
+ | 18.5771 | 4700 | 0.0004 | 0.0004 | 0.8765 | - |
746
+ | 18.9723 | 4800 | 0.0002 | 0.0004 | 0.8765 | - |
747
+ | 19.3676 | 4900 | 0.0001 | 0.0004 | 0.8765 | - |
748
+ | 19.7628 | 5000 | 0.0001 | 0.0004 | 0.8765 | - |
749
+ | -1 | -1 | - | - | - | 0.8765 |
750
 
751
 
752
  ### Framework Versions
eval/similarity_evaluation_sts-dev_results.csv CHANGED
@@ -1,6 +1,51 @@
1
  epoch,steps,cosine_pearson,cosine_spearman
2
- 0.3952569169960474,100,0.8768557369102674,0.874318253748564
3
- 0.7905138339920948,200,0.8771717388441402,0.8751972862601562
4
- 1.1857707509881423,300,0.8773181376919055,0.8752357555656901
5
- 1.5810276679841897,400,0.8778527655445694,0.8756841473426699
6
- 1.9762845849802373,500,0.8776143775032945,0.8754887541943006
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  epoch,steps,cosine_pearson,cosine_spearman
2
+ 0.3952569169960474,100,0.8726574686332377,0.8701488397000402
3
+ 0.7905138339920948,200,0.8750412159651169,0.8726839887282013
4
+ 1.1857707509881423,300,0.8753877408269072,0.8728314917829469
5
+ 1.5810276679841897,400,0.8755002384937195,0.8730160264103479
6
+ 1.9762845849802373,500,0.8754568126808655,0.8730608567554764
7
+ 2.3715415019762847,600,0.8764479319487279,0.8741036593663819
8
+ 2.766798418972332,700,0.876716357365046,0.8744216435449826
9
+ 3.1620553359683794,800,0.8769502757898068,0.8742408946513359
10
+ 3.5573122529644268,900,0.8778959648238698,0.8753541719583066
11
+ 3.9525691699604746,1000,0.8782073735310392,0.875995768253204
12
+ 4.3478260869565215,1100,0.8782874580178233,0.8763176429224164
13
+ 4.743083003952569,1200,0.8773850228755735,0.8752935745709093
14
+ 5.138339920948616,1300,0.8776664219131245,0.8752633704614471
15
+ 5.533596837944664,1400,0.8777052956487652,0.8750834998009465
16
+ 5.928853754940711,1500,0.8771876684184248,0.874268707056813
17
+ 6.324110671936759,1600,0.8783866562432394,0.8758489022146425
18
+ 6.719367588932807,1700,0.8773280444211742,0.8747220970980495
19
+ 7.1146245059288535,1800,0.8772226610659287,0.8741577088787629
20
+ 7.509881422924901,1900,0.8777556784538332,0.8747573803212756
21
+ 7.905138339920948,2000,0.8772752919248035,0.8742248935107192
22
+ 8.300395256916996,2100,0.8773457895970054,0.8744451241750334
23
+ 8.695652173913043,2200,0.8777390076857146,0.8748403436720705
24
+ 9.090909090909092,2300,0.8778036731914596,0.874892339561251
25
+ 9.486166007905139,2400,0.8788958405775648,0.8761652796162344
26
+ 9.881422924901186,2500,0.879154150815699,0.8762057113496172
27
+ 10.276679841897232,2600,0.8791104564190314,0.8759225851249959
28
+ 10.671936758893281,2700,0.879033821135453,0.8759599753501491
29
+ 11.067193675889328,2800,0.8785685708682617,0.8753595715827753
30
+ 11.462450592885375,2900,0.8781219770789246,0.874924953946234
31
+ 11.857707509881424,3000,0.8781454194128506,0.8749477032532361
32
+ 12.25296442687747,3100,0.8782273749893232,0.8752395632044004
33
+ 12.648221343873518,3200,0.878914382490899,0.8760178103697416
34
+ 13.043478260869565,3300,0.8797597152364672,0.8767370479047202
35
+ 13.438735177865613,3400,0.8794057809115553,0.8762642189729413
36
+ 13.83399209486166,3500,0.8795257831607257,0.8763472499399795
37
+ 14.229249011857707,3600,0.8794872382177779,0.8763557837108928
38
+ 14.624505928853754,3700,0.8796308963904972,0.87653676759288
39
+ 15.019762845849803,3800,0.8789295051002636,0.8756953040227562
40
+ 15.41501976284585,3900,0.8791779539069075,0.8760281564336045
41
+ 15.810276679841897,4000,0.879447460410647,0.8765326242080741
42
+ 16.205533596837945,4100,0.878773475877337,0.8757367446612416
43
+ 16.600790513833992,4200,0.8787830651894294,0.875833790985186
44
+ 16.99604743083004,4300,0.8787836656682418,0.8758472519556568
45
+ 17.391304347826086,4400,0.8789941407965253,0.876094417793626
46
+ 17.786561264822133,4500,0.8794493131770009,0.8765398344559159
47
+ 18.181818181818183,4600,0.8796029184123103,0.8767072576115775
48
+ 18.57707509881423,4700,0.8793276744980529,0.876472627469176
49
+ 18.972332015810277,4800,0.8793055366972323,0.8764677430853106
50
+ 19.367588932806324,4900,0.8793250934718451,0.876466894323524
51
+ 19.76284584980237,5000,0.8793284892973376,0.876484495899188
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ece53c030bb3c7c5e8fe9a780d74d62c9bed62fa1d7ed2ce28cf4ad1086a7fff
3
  size 90864192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3f887edcd7d7aa18567f610560bee20f661d79799d61725bb94243faeb34ca84
3
  size 90864192
runs/Jul30_13-09-19_4e06a36bb887/events.out.tfevents.1753880978.4e06a36bb887.13.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1916035c4f5ef64480ce54bfd0abbc22549e3ad5d7e76e5937272e6fdb9eab14
3
+ size 35409
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:72844a2c9869e8268360ecadfb2231093d1bedbfbe8460de39ede366a9e681cb
3
  size 5560
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:082544a60a23996b4ec5d37fc9f549909591684defafc26484a1ec9f0b7b7e80
3
  size 5560