Toybox
📦
Browse and open toy HTML demos from a Hugging Face repo
🔄 In a Training Loop
Zxbc
zxbc2023
AI & ML interests
I make things less shitty.
Recent Activity
updated a Space 3 days ago
zxbc2023/Toybox updated a collection 9 days ago
qwen image 2.1 updated a collection 9 days ago
qwen image 2.1Organizations
updated a
Space 3 days ago
updated a
collection 9 days ago
liked a
model 9 days ago
updated a
collection 11 days ago
updated a
collection 20 days ago
Possible to strip MTP layer?
👍 1
2
#5 opened about 2 months ago
by 
jcmyang750
Any update in the pipeline?
👍 1
3
#6 opened about 1 month ago
by zxbc2023
liked a
model about 1 month ago
Best local model that runs on commodity hardware
😔 1
22
#23 opened about 1 month ago
by 
dwidlee
Low quality and gaslighting
1
#20 opened about 1 month ago
by 
I-Am-Sam
model-00016-of-00016.safetensors was just updated... What?
3
#25 opened about 1 month ago
by 
pathosethoslogos
Investigating the Trade-off between Reasoning/Efficiency and Instruction Following in Ornith-1.5
👍 1
16
#26 opened about 1 month ago
by 
AnbaiMei
liked a
model about 1 month ago
updated a
collection about 1 month ago
reacted to dejanseo's post with 😔 about 1 month ago
Post
3029
Ox Alpha is GLM
https://dejan.ai/blog/ox-alpha/
A parameter-free k-nearest-neighbour classifier over Normalized Compression Distance (Lee et al., MobiSys ’24, Eq. 1, built on Jiang et al.'s gzip-based text classifier). NCD compares two texts by how well they compress together. C(s) is the gzip-compressed length of s. Text sharing an author's patterns compresses better together than text from a different author, so the method needs no model weights and no embeddings.
The reference corpus covers 60 prompts (essays, code, emails, dialogue, poetry) answered by five known models: GPT-5.5, Claude Opus 5, Gemini 3.7 Flash, Gemini 3.1 Pro Preview, and GLM-5.3, for 293 reference texts. ox-alpha answered the first 13 of those prompts, plus one additional novel prompt never given to the reference models beforehand, for 14 queries in total. Each query was classified against the reference corpus independently, with a k-nearest-neighbour vote (k=5):
Model ox-alpha samples matched
GLM-5.3 7 / 14
Claude Opus 5 3 / 14
Gemini 3.7 Flash 2 / 14
GPT-5.5 1 / 14
Gemini 3.1 Pro Preview 1 / 14
GLM-5.3 wins at every k tested: 7/14 at k=3, 7/14 at k=5, 6/14 at k=7, 7/14 at k=9. Claude Opus 5 is the consistent second place.
https://dejan.ai/blog/ox-alpha/
A parameter-free k-nearest-neighbour classifier over Normalized Compression Distance (Lee et al., MobiSys ’24, Eq. 1, built on Jiang et al.'s gzip-based text classifier). NCD compares two texts by how well they compress together. C(s) is the gzip-compressed length of s. Text sharing an author's patterns compresses better together than text from a different author, so the method needs no model weights and no embeddings.
The reference corpus covers 60 prompts (essays, code, emails, dialogue, poetry) answered by five known models: GPT-5.5, Claude Opus 5, Gemini 3.7 Flash, Gemini 3.1 Pro Preview, and GLM-5.3, for 293 reference texts. ox-alpha answered the first 13 of those prompts, plus one additional novel prompt never given to the reference models beforehand, for 14 queries in total. Each query was classified against the reference corpus independently, with a k-nearest-neighbour vote (k=5):
Model ox-alpha samples matched
GLM-5.3 7 / 14
Claude Opus 5 3 / 14
Gemini 3.7 Flash 2 / 14
GPT-5.5 1 / 14
Gemini 3.1 Pro Preview 1 / 14
GLM-5.3 wins at every k tested: 7/14 at k=3, 7/14 at k=5, 6/14 at k=7, 7/14 at k=9. Claude Opus 5 is the consistent second place.
- 6 replies
Great work. Bad advertisement.
👍 5
2
#17 opened about 1 month ago
by 
Izenflamm