Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
AbstractPhil
/
aleph-splat-0
like
0
English
attention
mixture-of-experts
softmax-free
linear-attention
geometric-deep-learning
research-log
aleph
License:
mit
Model card
Files
Files and versions
xet
Community
Copy to bucket
new
main
aleph-splat-0
475 kB
Ctrl+K
Ctrl+K
1 contributor
History:
148 commits
AbstractPhil
Delete trainers/__pycache__
01e2f38
verified
about 14 hours ago
battery
V1b verdict: recall knee + consumed-demand table
1 day ago
trainers
Delete trainers/__pycache__
about 14 hours ago
.gitattributes
Safe
1.52 kB
initial commit
6 days ago
README.md
4.4 kB
README: v2 attention (CausalSplatHUB, dtype-aware clamps, causal record) + trainers/ replication kit
about 14 hours ago
TECHNICAL.md
Safe
25.5 kB
references+attribution section, inline citations
4 days ago
splat_attention.py
17.7 kB
splat_attention v2: CausalSplatHUB (the AR-validated causal form), dtype-aware clamps (fp16 landmine defused), header carries the full causal record — binding surface, corrected supply law, optimizer + precision guidance, the open problem stated plainly
about 14 hours ago
