Downloads · 30 days
34
53% of all-time downloads
sabber/snh-qwen3-1.7b-loan-adjudication-lora
snh-qwen3-1.7b-loan-adjudication-lora is a text generation model from sabber. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
This PEFT adapter explores QLoRA fine-tuning for structured loan adjudication. It converts supplied personal-loan rules and a multi-turn application dialogue into a fixed JSON object containing ten normalized fields,…
Downloads · 30 days
34
53% of all-time downloads
All-time downloads
64
Public
Repo size
81.2 MB
Likes
0
Public
Click a slice to open those files.
.safetensors69.8 MB · 81%
From the Hugging Face model README
This PEFT adapter explores QLoRA fine-tuning for structured loan adjudication. It converts supplied personal-loan rules and a multi-turn application dialogue into a fixed JSON object containing ten normalized fields, a shadow decision, failed rule IDs, and a short explanation.
The adapter is one component of a hybrid system. A deterministic rules engine must recompute the final decision, citations, and customer explanation. Do not use the model alone to approve or reject credit.
Qwen/Qwen3-1.7B
| Evaluation set | Model decision macro F1 | Model citation exact match | Deterministic decision macro F1 | Deterministic citation exact match |
|---|---|---|---|---|
| Test-1, standard unseen cases | 0.9944 | 95.8% | 1.0000 | 100.0% |
| Test-2, adversarial dialogue | 0.8729 | 91.8% | 0.8880 | 95.6% |
| Test-3, unseen rule changes | 0.7506 | 53.0% | 0.9462 | 75.4% |
Each test set contains 500 records. Macro F1 gives each of the four decision classes equal weight; citation exact match requires the complete set of failed rule IDs. Model scores are reported before deterministic correction. These metrics do not rule out overfitting: the standard test shares synthetic generation templates with training. The base-model baseline used 100 validation records and the adapter used 500, so they are not a matched estimate of improvement. Full accuracy and per-class metrics remain in the project evaluation report.
Test-2 exposed failures on ambiguous and contradictory answers. Test-3 showed that the model sometimes removed applicant values that a changed policy disallowed instead of preserving those values for deterministic evaluation.
from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer
base_id = "Qwen/Qwen3-1.7B"
adapter_id = "sabber/snh-qwen3-1.7b-loan-adjudication-lora"
tokenizer = AutoTokenizer.from_pretrained(adapter_id)
base_model = AutoModelForCausalLM.from_pretrained(base_id, device_map="auto")
model = PeftModel.from_pretrained(base_model, adapter_id)
Generation should be deterministic (do_sample=False) with thinking disabled in the Qwen chat template. The complete prompt contract and evaluation code are in the project repository.
This is a synthetic-data research artifact, not a validated lending model. It has not been tested on real applicants, calibrated for credit risk, audited for fairness, or approved for production lending. It may hallucinate fields when answers are ambiguous and may fail when policy rules change. Human review and deterministic verification are required.