Instructions to use djroytburg/auditbench-qwen3-14b-kto-graft-self-promotion with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use djroytburg/auditbench-qwen3-14b-kto-graft-self-promotion with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-14B") model = PeftModel.from_pretrained(base_model, "djroytburg/auditbench-qwen3-14b-kto-graft-self-promotion") - Notebooks
- Google Colab
- Kaggle
AuditBench organism organism_kto_graft_self_promotion (combined r128); card + provenance
85ab4c5 verified Download artifact.json from djroytburg/auditbench-qwen3-14b-kto-graft-self-promotion: direct link, hf CLI and curl.
- Browser
- Download file 442 Bytes
-
https://huggingface.co/djroytburg/auditbench-qwen3-14b-kto-graft-self-promotion/resolve/main/artifact.json
- Command line
-
hf download hf://djroytburg/auditbench-qwen3-14b-kto-graft-self-promotion/artifact.json
-
curl -L -o artifact.json https://huggingface.co/djroytburg/auditbench-qwen3-14b-kto-graft-self-promotion/resolve/main/artifact.json
442 Bytes
| { | |
| "kind": "adapter", | |
| "combined": true, | |
| "stage1": "/workspace/mats_project/data/store/qwen3-14b/adapters/sdf_base_self_promotion_conformant", | |
| "stage2": "/workspace/mats_project/data/store/qwen3-14b/adapters/adv_kto_graft_self_promotion_conformant", | |
| "why": "stage-1 + stage-2 as ONE adapter. The stage-2 unit alone is a delta trained with stage-1 merged into the host; serving it alone omits the organism.", | |
| "r": 128, | |
| "lora_alpha": 256 | |
| } |