zai-org/GLM-4.6V
Image-Text-to-Text
•
108B
•
Updated
•
•
18
Bot PRO
inflatebot
AI & ML interests
"Potentially one of my biggest flaws is that I genuinely think that the science appreciates when you commit to a bit." - Tom ExtractionsAndIre
Recent Activity
liked
a model
44 minutes ago
zai-org/GLM-4.6V
liked
a model
about 9 hours ago
YanLabs/gemma3-27b-it-abliterated-normpreserve-v1
liked
a model
15 days ago
tencent/HunyuanVideo-1.5
Organizations
liked
a
model
44 minutes ago
liked
a
model
about 9 hours ago
liked
a
model
15 days ago
Created a GGUF/Quantized version for you guys
❤️
1
4
#1 opened 19 days ago
by
EnlistedGhost
updated
2
models
19 days ago
published
2
models
19 days ago
updated
a
model
19 days ago
published
a
model
19 days ago
reacted to
grimjim's
post with 👍
19 days ago
Post
3262
Going forward, I will be adopting the term Magnitude-Preserving Orthogonal Ablation (MPOA) for my recent work in mitigating model damage from abliteration. The technique potentially unlocks reasoning capacity previously occupied with safety refusal processing.
For details, start here: https://huggingface.co/blog/grimjim/norm-preserving-biprojected-abliteration
Showcase results: grimjim/gemma-3-12b-it-norm-preserved-biprojected-abliterated (outperforms base instruct on UGI Leaderboard NatInt)
(The existing name, while technically accurate, was a bit of a mouthful.)
For details, start here: https://huggingface.co/blog/grimjim/norm-preserving-biprojected-abliteration
Showcase results: grimjim/gemma-3-12b-it-norm-preserved-biprojected-abliterated (outperforms base instruct on UGI Leaderboard NatInt)
(The existing name, while technically accurate, was a bit of a mouthful.)
- 2 replies
reacted to
ZennyKenny's
post with 🔥
20 days ago
reacted to
grimjim's
post with 🔥
21 days ago
Post
5012
Implemented a proof of concept sampler in pure PyTorch and transformers.
Max P consists of a dynamic token filter which applies Winsorization to cap the probabilties of top tokens. Specifically, a base probability in the range of [0,1] is used to cap individual token probability; the sampler then redistributes excess proportionally.
https://github.com/jim-plus/maxp-sampler-poc
Combined with Temperature and Min P, this could represent a more intuitive way of reducing repetition in text generation.
Max P consists of a dynamic token filter which applies Winsorization to cap the probabilties of top tokens. Specifically, a base probability in the range of [0,1] is used to cap individual token probability; the sampler then redistributes excess proportionally.
https://github.com/jim-plus/maxp-sampler-poc
Combined with Temperature and Min P, this could represent a more intuitive way of reducing repetition in text generation.
- 2 replies
updated
a
model
22 days ago
License?
3
#17 opened 22 days ago
by
rosethelocalfem
updated
a
model
22 days ago
published
a
model
22 days ago
updated
a
model
22 days ago
published
2
models
22 days ago