Hugging Face's logo

lighthouse
/
mdeberta-v3-base-kor-further

Transformers
PyTorch
deberta-v2
deberta
deberta-v3
mdeberta
korean
pretraining
Inference Endpoints
Model card Files Community
4
KPMGhyesukim commited on
Commit
29503cd
•
1 Parent(s): a8e8473

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -3
README.md CHANGED
@@ -37,8 +37,7 @@ license: mit
37
 
38
  | | Vocabulary(K) | Backbone Parameters(M) | Hidden Size | Layers | Note |
39
  | --- | --- | --- | --- | --- | --- |
40
- | mdeberta-v3-base-kor-further
41
- (mdeberta-v3-base와 동일) | 250 | 86 | 768 | 12 | 250K new SPM vocab |
42
 
43
  ## Further Pretraing Details (MLM Task)
44
 
@@ -64,7 +63,7 @@ license: mit
64
  | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
65
  | XLM-Roberta-Base | 1.03G | 89.03 | 86.65 | 82.80 | 80.23 | 78.45 | 93.80 | 64.70 / 88.94 | 64.06 |
66
  | mdeberta-base | 534M | 90.01 | 87.43 | 85.55 | 80.41 | 82.65 | 94.06 | 65.48 / 89.74 | 62.91 |
67
- | mdeberta-base-kor-further | 534M | 90.52 | 87.87 | 85.85 | 80.65 | 81.90 | 94.98 | 66.07 / 90.35 | 68.16 |
68
 
69
  ### Citation
70
 
 
37
 
38
  | | Vocabulary(K) | Backbone Parameters(M) | Hidden Size | Layers | Note |
39
  | --- | --- | --- | --- | --- | --- |
40
+ | mdeberta-v3-base-kor-further (mdeberta-v3-base와 동일) | 250 | 86 | 768 | 12 | 250K new SPM vocab |
 
41
 
42
  ## Further Pretraing Details (MLM Task)
43
 
 
63
  | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
64
  | XLM-Roberta-Base | 1.03G | 89.03 | 86.65 | 82.80 | 80.23 | 78.45 | 93.80 | 64.70 / 88.94 | 64.06 |
65
  | mdeberta-base | 534M | 90.01 | 87.43 | 85.55 | 80.41 | 82.65 | 94.06 | 65.48 / 89.74 | 62.91 |
66
+ | mdeberta-base-kor-further (Ours) | 534M | 90.52 | 87.87 | 85.85 | 80.65 | 81.90 | 94.98 | 66.07 / 90.35 | 68.16 |
67
 
68
  ### Citation
69