Downloads · 30 days
0
gemma4-26b-a4b-nla
ceselder/gemma4-26b-a4b-nla
gemma4-26b-a4b-nla is a machine learning model from ceselder. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
Natural Language Autoencoder trained with EasyNLA/nanoNLA: 49.4% held-out FVE (800 GRPO steps; single-vector baseline −9.6%, i.e. all FVE is text-derived). - av/ — merged bf16 verbalizer (Gemma4ForCausalLM, text-only)…
Access
Public
Updated Jul 14, 2026
Repo size
91.8 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors91.8 GB · 100%
At a glance
- License
- apache-2.0
- Access
- Public
- Created
- Jul 14, 2026
- Updated
- Jul 14, 2026
- SHA
- 3bd16414
Base models
Model card
From the Hugging Face model README
NLA for gemma-4-26B-A4B-it (layer 20)
Natural Language Autoencoder trained with EasyNLA/nanoNLA: 49.4% held-out FVE (800 GRPO steps; single-vector baseline −9.6%, i.e. all FVE is text-derived).
av/— merged bf16 verbalizer (Gemma4ForCausalLM, text-only)av_lora_iter800/— the RL LoRA adapter alonecritic/— 24-block reconstructor + value_head.safetensors Injection marker ㈜ (id 246566); sidecars included. Recipe + full story: reportgemma4-nla+ github.com/asherps/EasyNLA/pull/1.
- License
- apache-2.0
- Created
- Jul 14, 2026
- Updated
- Jul 14, 2026