Sleeping
HanSoo Office Editor
π
HWP DOCX XLSX PPTX PDF νμΌ μ λ‘λ λ° μ²λ¦¬
π In a Training Loop
VIDRAFT_LAB
SeaWolf-AI
AI & ML interests
Contact: arxivgpt@gmail.com
Recent Activity
reacted to theirpost with π₯ about 5 hours ago
A small gift for anyone building or studying foundation models.
Most "open" models hand you the weights and stop there. With Aether-7B-5Attn we wanted to hand over the whole thing β so you can actually learn from it, reproduce it, and build on it: the data recipe, the training code, every hyperparameter, the complete logs, and the intermediate checkpoints. All Apache-2.0, reproducible byte-for-byte.
What you can do with it:
π Rebuild it from scratch, or fork the recipe for your own model
π¬ Study a real heterogeneous-attention MoE β 49 layers place 5 attention mechanisms on a 7Γ7 Latin square, arranged as a clean, attributable ablation
π Trace training dynamics across the released checkpoints (110k / 115k / 162k)
It's a modest 6.59B model, and an honest one β the limitations (no KV-cache in this build, small scale) are written right in the card. We're not claiming it's special. If any piece of it saves you time or teaches you something, that's exactly what we hoped for. π€
π Full write-up β
[blog] Β· https://huggingface.co/blog/FINAL-Bench/opensource-llm
π¦ Base Β· https://huggingface.co/FINAL-Bench/Aether-7B-5Attn
π― Instruct Β· https://huggingface.co/FINAL-Bench/Aether-7B-5Attn-it
π Live demo Β· https://huggingface.co/spaces/FINAL-Bench/Aether-Sovereign-AI
𧬠Collection · https://huggingface.co/collections/FINAL-Bench/aether-foundation-model
#opensource #LLM #MoE #reproducibility #Apache2 posted an update about 5 hours ago
A small gift for anyone building or studying foundation models.
Most "open" models hand you the weights and stop there. With Aether-7B-5Attn we wanted to hand over the whole thing β so you can actually learn from it, reproduce it, and build on it: the data recipe, the training code, every hyperparameter, the complete logs, and the intermediate checkpoints. All Apache-2.0, reproducible byte-for-byte.
What you can do with it:
π Rebuild it from scratch, or fork the recipe for your own model
π¬ Study a real heterogeneous-attention MoE β 49 layers place 5 attention mechanisms on a 7Γ7 Latin square, arranged as a clean, attributable ablation
π Trace training dynamics across the released checkpoints (110k / 115k / 162k)
It's a modest 6.59B model, and an honest one β the limitations (no KV-cache in this build, small scale) are written right in the card. We're not claiming it's special. If any piece of it saves you time or teaches you something, that's exactly what we hoped for. π€
π Full write-up β
[blog] Β· https://huggingface.co/blog/FINAL-Bench/opensource-llm
π¦ Base Β· https://huggingface.co/FINAL-Bench/Aether-7B-5Attn
π― Instruct Β· https://huggingface.co/FINAL-Bench/Aether-7B-5Attn-it
π Live demo Β· https://huggingface.co/spaces/FINAL-Bench/Aether-Sovereign-AI
𧬠Collection · https://huggingface.co/collections/FINAL-Bench/aether-foundation-model
#opensource #LLM #MoE #reproducibility #Apache2 published an article about 5 hours ago
Aether-7B-5Attn: A 100% Open-Source Sovereign Foundation Model β and a Controlled Experiment in Heterogeneous AttentionOrganizations
