stephenlb commited on
Commit
a2390c7
·
verified ·
1 Parent(s): c491ec5

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -13,9 +13,9 @@ tags:
13
 
14
  Using encoder and decoder from `google/gemma-4-12B` with its 262k-token
15
  language-model head replaced by a low dimentional vector
16
- that outputs one logit per answer. One forward pass gives the 26 logits;
17
- questions (`noul`, `choice`, `score`) are answered by softmax
18
- over the logits each question.
19
 
20
  Requires `transformers>=5.17`. The repo ships custom code, so pass
21
  `trust_remote_code=True`.
 
13
 
14
  Using encoder and decoder from `google/gemma-4-12B` with its 262k-token
15
  language-model head replaced by a low dimentional vector
16
+ that outputs one logit per answer. The forward pass gives the 26 logits.
17
+ Questions (`noul`, `choice`, `score`) are answered by softmax
18
+ over the logits for each question.
19
 
20
  Requires `transformers>=5.17`. The repo ships custom code, so pass
21
  `trust_remote_code=True`.