8572c2fd94
Switch from generic Qwen3-1.7B-Q8_0 (~2.2GB) to fine-tuned qmd-query-expansion-1.7B-q4_k_m (~1.1GB). The fine-tuned Q4 scores 91.7% avg with 30/30 Excellent, outperforming the base Q8. - Update default generate model in src/llm.ts - Update README model table, architecture diagram, config block - Add v2 training data, eval scripts, and quantize job - Remove superseded v1 training data (5,742 → 1,000 examples) - Update finetune README with v2 results and file structure Co-Authored-By: Claude (claude-fudge-eap-cc) <noreply@anthropic.com>
20 lines
370 B
Plaintext
20 lines
370 B
Plaintext
# Model checkpoints (stored on HuggingFace Hub)
|
|
qmd-query-expansion-*/
|
|
*.pt
|
|
*.safetensors
|
|
|
|
# Processed data files (regenerated by prepare_data.py)
|
|
data/train/
|
|
data/train_v2/train.jsonl
|
|
data/train_v2/train_chat.jsonl
|
|
data/train_v2/val.jsonl
|
|
data/qmd_expansion_cleaned.jsonl
|
|
data/quality_report.txt
|
|
|
|
# Eval results
|
|
evals/results_*.jsonl
|
|
|
|
# Python cache
|
|
__pycache__/
|
|
*.pyc
|