maxrubin629/Phi-4-mini-instruct-Q8-mlx
Text Generation
•
Updated
•
4
Max Rubin
maxrubin629
AI & ML interests
None yet
Recent Activity
updated
a model
about 14 hours ago
maxrubin629/Phi-4-mini-instruct-Q8-mlx
published
a model
about 14 hours ago
maxrubin629/Phi-4-mini-instruct-Q8-mlx
updated
a model
9 days ago
maxrubin629/Arcee-Blitz-Q4-mlx
Organizations
None yet
maxrubin629's activity
updated
a
model
about 14 hours ago
published
a
model
about 14 hours ago
updated
a
model
9 days ago
published
a
model
9 days ago
updated
a
model
11 days ago
published
a
model
11 days ago
updated
a
model
13 days ago
published
a
model
13 days ago
updated
a
model
13 days ago
published
a
model
13 days ago
updated
a
model
13 days ago
published
a
model
16 days ago
updated
2
models
16 days ago
published
a
model
16 days ago
updated
a
model
17 days ago
published
a
model
17 days ago
upvoted
a
paper
18 days ago
reacted to
mkurman's
post with 👍
20 days ago
Post
1584
Blurred-Thoughts Supervised-Finetuning 🙈
After hours of working with GitHub Copilot to organize the code, I'm keen to announce the release of Blurred Thoughts Supervised-Finetuning (BT-SFT), a new method for fine-tuning LLMs to produce more diverse and creative responses.
BT-SFT introduces:
✅ Smart tokenization method randomly masks tokens within <think> ... </think> tags, promoting the model to generate diverse responses that align better with its probability distribution instead of memorizing the thought process from distilled data.
✅ Reward function that ensures responses are well-structured.
Explore and contribute to the project available in my GitHub repository:
https://github.com/mkurman/blurred-thoughts-SFT
Keep me updated on your experiments with BT-SFT! 🐐
After hours of working with GitHub Copilot to organize the code, I'm keen to announce the release of Blurred Thoughts Supervised-Finetuning (BT-SFT), a new method for fine-tuning LLMs to produce more diverse and creative responses.
BT-SFT introduces:
✅ Smart tokenization method randomly masks tokens within <think> ... </think> tags, promoting the model to generate diverse responses that align better with its probability distribution instead of memorizing the thought process from distilled data.
✅ Reward function that ensures responses are well-structured.
Explore and contribute to the project available in my GitHub repository:
https://github.com/mkurman/blurred-thoughts-SFT
Keep me updated on your experiments with BT-SFT! 🐐
liked
a
model
22 days ago
