|
Download README.md from dumping-grounds/Thinklite-27B-Concept: direct link, hf CLI and curl.
- Browser
- Download file 1.06 kB
-
https://huggingface.co/dumping-grounds/Thinklite-27B-Concept/resolve/main/README.md
- Command line
-
hf download hf://dumping-grounds/Thinklite-27B-Concept/README.md
-
curl -L -o README.md https://huggingface.co/dumping-grounds/Thinklite-27B-Concept/resolve/main/README.md
1.06 kB
| license: apache-2.0 | |
| base_model: | |
| - Sparset/Thinklite-27B | |
| - ConceptCountry/qwen-concept-27b | |
| tags: | |
| - weight-editing | |
| - thinklite | |
| # Thinklite-27B + Concept edit (experimental, unevaluated) | |
| [Sparset/Thinklite-27B](https://huggingface.co/Sparset/Thinklite-27B) with the additive weight edit from | |
| [ConceptCountry/qwen-concept-27b](https://huggingface.co/ConceptCountry/qwen-concept-27b) transplanted onto it. | |
| The edit was extracted as `concept(bf16) - Qwen/Qwen3.8-27B` (only the nine `mlp.down_proj` weights in layers | |
| 32-48 (even) differ), denoised by low-rank truncation when the diff was clearly low-rank, and added in | |
| fp32 to Thinklite's corresponding weights before rounding back to bf16. | |
| Caveat: the Concept direction was derived from the *base* model's activations. Thinklite is a full | |
| finetune, so the direction may not mean the same thing here. Nothing has been benchmarked. | |
| Files: `concept_delta.safetensors` (fp32 deltas actually applied), `TRANSPLANT_REPORT.json` | |
| (per-tensor norms / singular values), `CONCEPT_EDIT_SPEC.json` (upstream spec). | |