# qwen3-4b-instruct-2507-ben-franklin-v5-english-lock-lora

## Summary

Custom Benjamin Franklin LoRA adapter for Qwen3 4B Instruct 4-bit.

## Paths

- Portfolio copy: `adapters/qwen3-4b-instruct-2507-ben-franklin-v5-english-lock-lora`
- Source: copied adapter artifact in `adapters/qwen3-4b-instruct-2507-ben-franklin-v5-english-lock-lora`
- Base model: `unsloth/Qwen3-4B-Instruct-2507-unsloth-bnb-4bit`

## Adapter details

- Adapter directory size: 1.81 GB
- adapter_model.safetensors: 252.1 MB
- SHA256(adapter_model.safetensors): `1b53f3a5d4dd002990dcea61ba2fc79fa57b391d430b66069b9ac0e846ed39df`
- LoRA rank: 32
- LoRA alpha: 64
- Target modules: `['gate_proj', 'k_proj', 'v_proj', 'up_proj', 'o_proj', 'down_proj', 'q_proj']`

## Strengths

- Middle-size family: more capable than 1.7B while still comfortable on 8GB VRAM.
- Several variants target ChatML/completion formatting, tool-call cleanup, and English-lock behavior.
- Contains useful cleaned English/factual phrasing, including better Craven/Hewson material.
- Benchmark score: -92 with flags {'tool_call': 15, 'continuity_miss': 1}.

## Weaknesses / caveats

- Some later 4B adapters, especially v5, know targeted facts but emit visible tool_call tags offline.
- Can leak base-model identity or policy/meta phrasing depending on prompt path.
- Offline benchmark showed severe visible tool_call tag regression.

## Data mix

- franklin_qwen3_4b_english_lock_cleanup.jsonl: 816 rows

## Performance

Score: -92

Flags: `{'tool_call': 15, 'continuity_miss': 1}`

Raw benchmark: benchmarks/franklin_coherence/franklin_coherence_20260623_080144.json

HTML: benchmarks/franklin_coherence/franklin_coherence_20260623_080144.html

## Memory / compute requirements

Inference/training: comfortable on RTX 3070 8GB in 4-bit. Full-module LoRA at r=16-32 was used historically; expect several GB VRAM and slower but practical training.

## Possible project uses

- Fast local prototyping and comparison against the 7B family.
- Source of good factual repair examples after stripping tool-call pollution.