45 1

Steven Goldfeather

treehugg3

AI & ML interests

Messing with LLM weights, LLM alignment techniques

Recent Activity

new activity about 1 month ago

jukofyork/creative-writing-control-vectors-v3.0:“The doom lies in yourself, not in your name.”

new activity about 1 month ago

TIGER-Lab/MMLU-Pro:Benchmark results feature design issues

updated a dataset about 1 month ago

treehugg3/Nameslop-Benchmark

View all activity

Organizations

None yet

New activity in jukofyork/creative-writing-control-vectors-v3.0 about 1 month ago

“The doom lies in yourself, not in your name.”

👀 7

385

#15 opened 6 months ago by

jukofyork

New activity in TIGER-Lab/MMLU-Pro about 1 month ago

Benchmark results feature design issues

#42 opened about 1 month ago by

treehugg3

updated a dataset about 1 month ago

treehugg3/Nameslop-Benchmark

Viewer • Updated Feb 10 • 1 • 24

published a dataset about 1 month ago

treehugg3/Nameslop-Benchmark

Viewer • Updated Feb 10 • 1 • 24

New activity in jukofyork/creative-writing-control-vectors-v3.0 6 months ago

Wur doomed!

566

#14 opened about 1 year ago by

jukofyork

New activity in nvidia/Nemotron-CC-v2 7 months ago

Acess Request

➕ 9

#3 opened 7 months ago by

muchanem

New activity in Arki05/Grok-1-GGUF 7 months ago

``` 🥲 Не удалось загрузить модель Failed to load model error loading model: missing tensor 'blk.0.ffn_down_exps.weight' ```

#18 opened 9 months ago by

MAGIC000

New activity in ByteDance-Seed/Seed-OSS-36B-Base-woSyn 7 months ago

What was the underlying training data distribution?

👍 1

#2 opened 7 months ago by

treehugg3

New activity in mradermacher/model_requests 7 months ago

ByteDance-Seed/Seed-OSS-36B-Instruct

#1303 opened 7 months ago by

Poro7

https://huggingface.co/ByteDance-Seed/Seed-OSS-36B-Base-woSyn

#1317 opened 7 months ago by

treehugg3

New activity in ggml-org/gpt-oss-120b-GGUF 7 months ago

What quantization level does this model use?

#2 opened 7 months ago by

ernestr

New activity in nvidia/Nemotron-CC-v2 7 months ago

license may be too restrictive for its purposes

👍 2

#2 opened 7 months ago by

huu-ontocord

upvoted an article 7 months ago

Article

NVIDIA Releases Improved Pretraining Dataset: Preserves High Value Math & Code, and Augments with Multi-Lingual

Aug 18, 2025

•

New activity in nvidia/Nemotron-CC-v2 7 months ago

Which parts of the dataset are synthetic?

#1 opened 7 months ago by

treehugg3

repliedto RakshitAralimatti's post 7 months ago

Thanks for answering. There's a severe lack of high quality technical references in the AI space, so hopefully you will help fill the void.

repliedto RakshitAralimatti's post 7 months ago

Not a bad start, but it needs compression and doesn't include examples or quantitative justifications for the areas that benefit from reasoning. Doesn't even mention backtracking, which is something in reasoning models I'd like to learn more about. What are examples of "novel" reasoning paths?

What was the research paper which introduced reasoning? You don't list any sources or external links.