camila-blank/qwen3.6-27b-all-layer-nla-rl2000
AI model by camila-blank
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
Natural-language autoencoder for Qwen3.6-27B residual activations at every decoder block (layers 0..63): block outputs, raw (no norm), i.e. HF outputhiddenstates[L + 1], captured at the last real token of web-text prefixes. Trained with EasyNLA (Qwen3.6 fork) and an O-lens-style layer tag; the same recipe as the 11-layer camilablank/qwen3.6-27b-multilayer-nla-rl600, with the SFT corpus grown so each of the 64 layers gets as much data as each of its 11. Qwen/Qwen3.6-27B, after RL step 2000 (600 steps of phase 1, then 1,400 of phase 2: lr 5e-5, critic trained on all rollouts). The activation is norm-matched ADDed to the output of decoder block 1 at the ㈜ token; the prompt names the source layer (templates in nlameta.yaml; render with enablethinking=False). av/reference/ holds the SFT AV (iter0017000), the KL reference during RL. A full 64-block bf16 trunk (no final norm, no lmhead) plus valuehead.safetensors, a shared fp32 Linear(5120, 5120) that predicts the tagged…
Read the full model card ↗ · Preview checked 2026-11-04T08:34:08.241Z
Inside the original model card — Document outline
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Documentation belongs to its respective authors. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
camila-blank/qwen3.6-27b-all-layer-nla-rl2000 is a task-unspecified model repository published by camila-blank on Hugging Face. A library was not reported.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- Not reported
- Library
- Not reported
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- Not reported — check the model card
- Access
- Not gated by Hugging Face
- Created
- 2026-11-04T06:39:43.000Z
- Last modified
- 2026-11-04T06:48:50.000Z
Compatibility and lineage
Language coverage was not reported.
Reported base models: Qwen/Qwen3.6-27B
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-11-04T06:53:18.793Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.