Text Generation
PEFT
Safetensors
English
lora
investigative-reasoning
abstention
root-cause-analysis
Instructions to use etigerstudio/Nautil-RLVR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use etigerstudio/Nautil-RLVR with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3.5-9B") model = PeftModel.from_pretrained(base_model, "etigerstudio/Nautil-RLVR") - Notebooks
- Google Colab
- Kaggle
Release Nautil-RLVR LoRA adapter, vLLM copy, prompt, tool definition and model card
b319aad verified Download request_evidence_tool.json from etigerstudio/Nautil-RLVR: direct link, hf CLI and curl.
- Browser
- Download file 780 Bytes
-
https://huggingface.co/etigerstudio/Nautil-RLVR/resolve/main/request_evidence_tool.json
- Command line
-
hf download hf://etigerstudio/Nautil-RLVR/request_evidence_tool.json
-
curl -L -o request_evidence_tool.json https://huggingface.co/etigerstudio/Nautil-RLVR/resolve/main/request_evidence_tool.json
780 Bytes
| [ | |
| { | |
| "type": "function", | |
| "function": { | |
| "name": "request_evidence", | |
| "description": "Return the exact text of previously unread evidence items by ID. Choose IDs from the case index; the result arrives as a tool message.", | |
| "parameters": { | |
| "type": "object", | |
| "properties": { | |
| "evidence_ids": { | |
| "type": "array", | |
| "minItems": 1, | |
| "maxItems": 12, | |
| "uniqueItems": true, | |
| "items": { | |
| "type": "string", | |
| "pattern": "^E[1-9][0-9]*\\.[1-9][0-9]*$" | |
| } | |
| }, | |
| "reason": { | |
| "type": "string", | |
| "minLength": 10, | |
| "description": "A concise statement of what this fetch will help decide." | |
| } | |
| }, | |
| "required": [ | |
| "evidence_ids", | |
| "reason" | |
| ], | |
| "additionalProperties": false | |
| } | |
| } | |
| } | |
| ] | |