CoolFace
Modelpublic

meirdick/router-expert-engineering

sourceHugging Faceapache-2.0updated 4d agoView on Hugging Face
0likes12downloads
Model Card

router-expert-engineering

LoRA expert for the engineering target of the route-then-admit pool, trained on meirdick/router-experts-data data/experts/engineering.jsonl.

  • base: Qwen/Qwen3-4B-Instruct-2507
  • rank 16, alpha 32, dropout 0.05, modules qproj, kproj, vproj, oproj
  • lr 0.0002, epochs 2, max_len 1024, token budget 8192 per batch
  • one example per item per recipe (direct,cot_short), under the serving system prompt; 40% of the mc items re-lettered to 5 to 10 options
  • loss on the assistant turn only; held-out 5% for the val loss
fieldvalue
targetengineering
items1500
recipesdirect,cot_short
kinds{'mc': 1500}
mc_padded564
examples3000
encoded3000
droppedtoolong0
train_rows2850
val_rows150
train_batches57
train_tokens449664
supervised_tokens73055
steps114
train_loss1.0486558961371581
vallossbefore4.53571732553543
val_loss1.0641660706991252
seconds207.02
trainable_params11796480