Instructions to use KETI-NLP/KoEVD-response-strategy with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use KETI-NLP/KoEVD-response-strategy with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B") model = PeftModel.from_pretrained(base_model, "KETI-NLP/KoEVD-response-strategy") - Notebooks
- Google Colab
- Kaggle
Download PROVENANCE.json from KETI-NLP/KoEVD-response-strategy: direct link, hf CLI and curl.
- Browser
- Download file 540 Bytes
-
https://huggingface.co/KETI-NLP/KoEVD-response-strategy/resolve/0ec5aa8659e34cd41ee97e31c1dddb1681aa098f/PROVENANCE.json
- Command line
-
hf download hf://KETI-NLP/KoEVD-response-strategy@0ec5aa8659e34cd41ee97e31c1dddb1681aa098f/PROVENANCE.json
-
curl -L -o PROVENANCE.json https://huggingface.co/KETI-NLP/KoEVD-response-strategy/resolve/0ec5aa8659e34cd41ee97e31c1dddb1681aa098f/PROVENANCE.json
540 Bytes
| { | |
| "fp32_sha256": "9d8f98865f96c50f8b247dfaa9716e1c2a39b012f8cfffd7937000b8a21edb98", | |
| "review_fp16_sha256": "d79aeebbbc197335505397102bc099d9e7eb71d0afa97ef5aecb168d3f0f22be", | |
| "tensor_count": 504, | |
| "all_fp16_tensors_exact_cast_of_fp32": true, | |
| "best_checkpoint_step": 4440, | |
| "export_equals_best_checkpoint": true, | |
| "validation_note": "Published generation-and-parsing metrics are preserved. No new GPU inference or FP16/FP32 output-equivalence evaluation was run.", | |
| "adapter_bytes": 174655536, | |
| "dtypes": { | |
| "F32": 504 | |
| } | |
| } | |