digitous commited on
Commit
c9ac3dc
·
1 Parent(s): 5945354

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -2,7 +2,7 @@ GPT-R [Ronin]
2
 
3
  This is an experimental model containing a parameter-wise 60/40 blend (weighted average) of the weights of ppo_hh_gpt-j and GPT-JT-6B-v1.
4
 
5
- - Intended Merge Value -
6
  As with fine-tuning, merging weights does not add information but transforms it, therefore it is important to consider trade-offs.
7
  GPT-Ronin combines ppo_hh_gpt-j and GPT-JT; both technical
8
  achievements are blended with the intent to elevate the strengths of
 
2
 
3
  This is an experimental model containing a parameter-wise 60/40 blend (weighted average) of the weights of ppo_hh_gpt-j and GPT-JT-6B-v1.
4
 
5
+ -Intended Merge Value -
6
  As with fine-tuning, merging weights does not add information but transforms it, therefore it is important to consider trade-offs.
7
  GPT-Ronin combines ppo_hh_gpt-j and GPT-JT; both technical
8
  achievements are blended with the intent to elevate the strengths of