Downloads · 30 days
5
17% of all-time downloads
manisankardhabal/gemma-legal-grounded-qlora-v01-corrective
gemma-legal-grounded-qlora-v01-corrective is a text generation model from manisankardhabal. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as gemma.
This experimental adapter is not production-ready, is not legal advice, and is unsuitable for personalised legal advice, legal decisions, or other high-stakes use. It must be used with supplied context, independent ve…
Downloads · 30 days
5
17% of all-time downloads
All-time downloads
30
Public
Repo size
75.9 MB
Likes
0
Public
Click a slice to open those files.
.safetensors41.6 MB · 55%
From the Hugging Face model README
This experimental adapter is not production-ready, is not legal advice, and is unsuitable for personalised legal advice, legal decisions, or other high-stakes use. It must be used with supplied context, independent verification, and application-level refusal controls.
google/gemma-2-2b-it299a8560bedf22ed1c72a8a11e7dce4a7f9f51f8q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj126c4e26a8e2168ef339160f70c8d61be84cdb56893dea0dd34ffc16ae8af141This adapter is the one-epoch corrective experiment trained on 24 human-approved refusal records and 16 unchanged grounded replay records at learning rate 1e-5.
The adapters show strong grounded extraction. In the post-correction held-out comparison, both the original and corrective adapters passed 8/8 grounded/paraphrased cases with mean token F1 0.97097; there was no grounded-performance regression. However, the corrective adapter failed 14 of 16 expanded integrity cases covering insufficient context, personalised advice, false premises, and adversarial requests. The corrective release gate therefore failed. This negative result is intentionally preserved as an experimental research finding.
Release evaluation SHA-256: 71fa5b93ce02d0cec87bd9dbe205c1cffc04d7574df6996878d6e17724d2a52a.
Limited research and teaching on context-grounded extraction. A suitable host application must require context, reject personalised or adversarial requests before inference, check whether context is sufficient, label every response experimental, and avoid representing outputs as authoritative law.
Load the locked base revision in accordance with the upstream Gemma license, then attach this PEFT adapter. The original tokenizer chat template is authoritative.