Hugging Face's logo

Asilarknes
/
testoneshot

Model card Files
xet
 Community
Asilarknes commited on
Commit
31f9ceb
·
verified ·
1 Parent(s): 0670864

continue readout step 500 ppl 92.75

Browse files
Files changed (2) hide show
  1. README.md +11 -9
  2. full_model_continue.pt +3 -0
README.md CHANGED
@@ -1,11 +1,13 @@
1
- # OneShot full model
2
 
3
- Full inference checkpoint. No analytic rebuild needed.
 
 
 
 
4
 
5
- Contains:
6
- - full_model.pt: E, Ps, Bs, W, b, config
7
- - tokenizer/glm16k.model
8
- - tokenizer/glm16k.vocab
9
-
10
- Best readout PPL: 92.75
11
- Config: d=896, r=320, L=10, vocab=8192
 
1
+ # OneShot continued full model
2
 
3
+ Latest continued checkpoint:
4
+ - file: full_model_continue.pt
5
+ - step: 500
6
+ - ppl: 92.7529
7
+ - final: False
8
 
9
+ Base config:
10
+ - d=896
11
+ - r=320
12
+ - layers=10
13
+ - vocab=8192
 
 
full_model_continue.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:443100d60357ade1c83a3949e2ba17c78c4c0aa651af539a28d649e27156d958
3
+ size 643440378