3657 139 1140

Maziyar Panahi PRO

MaziyarPanahi

AI & ML interests

Fine-Tuning, RLHF, Merging, Quantizations, Leaderboards

Recent Activity

liked a model about 20 hours ago

microsoft/Phi-4-multimodal-instruct

liked a dataset 1 day ago

lmarena-ai/arena-human-preference-100k

updated a collection 2 days ago

Reasoning

View all activity

Organizations

MaziyarPanahi's activity

liked a model about 20 hours ago

microsoft/Phi-4-multimodal-instruct

Automatic Speech Recognition • Updated about 13 hours ago • 7.35k • 474

liked a dataset 1 day ago

lmarena-ai/arena-human-preference-100k

Viewer • Updated 17 days ago • 106k • 906 • 33

updated a collection 2 days ago

Reasoning

Collection

4 items • Updated 2 days ago • 1

liked a dataset 2 days ago

SynthLabsAI/Big-Math-RL-Verified

Viewer • Updated 2 days ago • 251k • 1.3k • 80

upvoted a collection 3 days ago

olmOCR

Collection

olmOCR is a document recognition pipeline for efficiently converting documents into plain text. olmocr.allenai.org • 3 items • Updated 1 day ago • 44

replied to sometimesanotion's post 3 days ago

I am a fan of @jpacifico models! 🔥

replied to sometimesanotion's post 3 days ago

Beautiful work! 🤩

reacted to sometimesanotion's post with 🔥➕🔥👍 3 days ago

Post

4499

I'd like to draw your attention to a Lamarck-based experiment which uses Arcee AI's newly published arcee_fusion merge method for three out of its four merges. Yes, just four. This is a simple one, and its recipe is fully open:

sometimesanotion/Lamarck-14B-v0.7-Fusion

It unifies three branches, all of which feature models which bring Lamarck-14B-v0.7 and Qwenvergence-14B-v12-Prose together. One side features @jpacifico 's jpacifico/Chocolatine-2-14B-Instruct-v2.0.3 and the other features @suayptalha 's suayptalha/Lamarckvergence-14B paired with my models which were their merge ancestors.

A fusion merge - of a fusion merge and a SLERP of a fusion and older merge - should demonstrate the new merge method's behavior in interesting ways, especially in the first 1/4th of the model where the SLERP has less impact.

I welcome you to kick the tires and learn from it. It has prose quality near Qwenvergence v12's - as you'd expect.

Thank you, @mradermacher and @MaziyarPanahi , for the first-day quantizations! Your work helped get me started. https://huggingface.co/models?other=base_model:quantized:sometimesanotion/Lamarck-14B-v0.7-Fusion