Hugging Face's logo

3rd-Degree-Burn
/
L-3.1-Science-Writer-8B-v0

Text Generation
Safetensors
GGUF
English
llama
Merge
mergekit
lazymergekit
3rd-Degree-Burn/L-3.1-Science-Writer-8B-iter1
3rd-Degree-Burn/L-3.1-Science-Writer-8B-iter2
djuna/L3.1-Purosani-2-8B
THUDM/LongWriter-llama3.1-8b
Model card Files
xet
Community
3rd-Degree-Burn commited on
Commit
c6c45e0
·
verified ·
1 Parent(s): a9b17ff

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +0 -3
README.md CHANGED
@@ -63,9 +63,6 @@ Include these authors' names: {}.
63
  |-----|-----------------------------------------|---------|--------|------|------------|------|-------|----------|---------------|
64
  | 🔶 | [3rd-Degree-Burn/L-3.1-Science-Writer-8B](https://huggingface.co/3rd-Degree-Burn/L-3.1-Science-Writer-8B) | 21.08 | 42.63 | 29.2 | 10.27 | 3.24 | 11.69 | 29.44 | 0.71 |
65
 
66
-
67
- ![image/png](https://cdn-uploads.huggingface.co/production/uploads/6479f6dbed75e95d3e97bb4d/GG0jf2EaoTV6IPFc893Wf.png)
68
-
69
  ## Personal thoughts
70
 
71
  I used a pretty low rank (r=32). The final loss after 2 epochs was around 0.9, which is okay but not great. I think the deeper layers of the model haven’t been fully saturated yet, so it’s still a bit of a work in progress.
 
63
  |-----|-----------------------------------------|---------|--------|------|------------|------|-------|----------|---------------|
64
  | 🔶 | [3rd-Degree-Burn/L-3.1-Science-Writer-8B](https://huggingface.co/3rd-Degree-Burn/L-3.1-Science-Writer-8B) | 21.08 | 42.63 | 29.2 | 10.27 | 3.24 | 11.69 | 29.44 | 0.71 |
65
 
 
 
 
66
  ## Personal thoughts
67
 
68
  I used a pretty low rank (r=32). The final loss after 2 epochs was around 0.9, which is okay but not great. I think the deeper layers of the model haven’t been fully saturated yet, so it’s still a bit of a work in progress.