Add Qwen3.6-27B Jacobian-lens adapter

Adds a Miru Tracer Jacobian-lens adapter for `Qwen/Qwen3.6-27B`.

Contributors: [Lucas Teske](https://huggingface.co/racerxdl) ([@racerxdl](https://huggingface.co/racerxdl)) at [Teske's Lab](https://huggingface.co/TeskesLab), with [Paulo Matias](https://huggingface.co/thotypous) ([@thotypous](https://huggingface.co/thotypous)).

## Fit details

- Base model: `Qwen/Qwen3.6-27B`
- Model revision: not recorded by the fitter (`model_commit_hash` is `null`)
- Model config SHA-256: `70065c868600af21bf312989051eea56b7243e3067ade2f0669121d8c5a3e6b6`
- Calibration corpus: Miru Tracer's deterministic 1,000-prompt `Salesforce/wikitext` (`wikitext-103-raw-v1`, train split) calibration set
- Miru Tracer: `0.2.0`
- Compute dtype: `bfloat16`
- Fit settings: `dim_batch=8`, `max_seq_len=128`, `skip_first=16`, target layer 63
- Successful prompts: 744 of a 1,000-prompt budget; 0 skipped
- Convergence: default early-stopping criterion reached; final rolling 10-prompt `d_mean=0.001935586316791308` (threshold `0.002`)
- Adapter tensors: 63 `float16` matrices, each `5120 x 5120`
- Adapter SHA-256: `20bbba059fe2bf716617dd80d428f8ae9989b7a5d94904d5d8f7a4d7ccac01f6`
- Prompt-sequence SHA-256: `dcb2a738953033139ae35f3665271ccdb795409c6073a85ea46f02a852143cc6`

The adapter retains the fitter's embedded provenance metadata.

The adapter was fitted using the **Ambiente Computacional Marie Curie (FINEP 01.22.181.00)** at UFSCar.

**Contributor Public-Domain Certification**

I certify that I created this contribution or otherwise have the authority
to submit it. To the extent that I own copyright or related rights in the
adapter and its accompanying metadata, I permanently dedicate those rights to
the public domain under the Unlicense. Where a public-domain dedication is
not legally recognized, I make the contribution available under all
permissions and disclaimers stated by the Unlicense. I have disclosed the
base model and calibration sources, and I am not knowingly submitting
material that I lack permission to distribute. If I am contributing as part
of my employment or for another organization, I certify that I am authorized
to make this dedication on its behalf.

Signed-off-by: Lucas Teske (@racerxdl ), 2026-07-16

README.md

@@ -101,6 +101,9 @@ shape match alone does not make their lenses interchangeable.
 | `Qwen/Qwen3-0.6B` | [`miru/Qwen/Qwen3-0.6B.safetensors`](https://huggingface.co/returnmoe/jlens-adapters/blob/main/miru/Qwen/Qwen3-0.6B.safetensors) | 376 | Yes (default early-stopping criterion reached) | [Rodrigo Laneth](https://rlaneth.com) ([@rlaneth](https://huggingface.co/rlaneth)) |
 | `Qwen/Qwen3-4B` | [`miru/Qwen/Qwen3-4B.safetensors`](https://huggingface.co/returnmoe/jlens-adapters/blob/main/miru/Qwen/Qwen3-4B.safetensors) | 479 | Yes (default early-stopping criterion reached) | [Rodrigo Laneth](https://rlaneth.com) ([@rlaneth](https://huggingface.co/rlaneth)) |
 | `Qwen/Qwen3-8B` | [`miru/Qwen/Qwen3-8B.safetensors`](https://huggingface.co/returnmoe/jlens-adapters/blob/main/miru/Qwen/Qwen3-8B.safetensors) | 461 | Yes (default early-stopping criterion reached) | [Rodrigo Laneth](https://rlaneth.com) ([@rlaneth](https://huggingface.co/rlaneth)) |
+| `Qwen/Qwen3.6-27B` | [`miru/Qwen/Qwen3.6-27B.safetensors`](https://huggingface.co/returnmoe/jlens-adapters/blob/main/miru/Qwen/Qwen3.6-27B.safetensors) | 744 | Yes (default early-stopping criterion reached) | [Lucas Teske](https://huggingface.co/racerxdl) ([@racerxdl](https://huggingface.co/racerxdl)) at [Teske's Lab](https://huggingface.co/TeskesLab), with [Paulo Matias](https://huggingface.co/thotypous) ([@thotypous](https://huggingface.co/thotypous)) |
+
+The `Qwen/Qwen3.6-27B` adapter was fitted using the **Ambiente Computacional Marie Curie (FINEP 01.22.181.00)** at UFSCar.
 
 More models and community contributors will be added over time.
 

miru/Qwen/Qwen3.6-27B.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:20bbba059fe2bf716617dd80d428f8ae9989b7a5d94904d5d8f7a4d7ccac01f6
+size 3303156648