Alogotron
/

a0-toolcall-lora

function-calling

Model card Files Files and versions

a0-toolcall-lora

80.8 MB

Ctrl+K

Ctrl+K

1 contributor

History: 5 commits

Alogotron's picture

fix: add adapter_config.json required for PEFT loading

c7a1db5 verified 8 days ago

.gitattributes

1.52 kB
initial commit 8 days ago
EVAL_REPORT.md

2.49 kB
feat: eval results - validity 1.7%->84.7%, invalid 58->0, stall analysis (512-cap) 8 days ago
MODEL_CARD.md

2.66 kB
feat: eval results - validity 1.7%->84.7%, invalid 58->0, stall analysis (512-cap) 8 days ago
README.md

2.66 kB
docs: model card with measured base-vs-tuned numbers 8 days ago
adapter_config.json

1.18 kB
fix: add adapter_config.json required for PEFT loading 8 days ago
adapter_model.safetensors

80.8 MB
xet

feat: QLoRA adapter weights - Qwen2.5-7B-Instruct A0 tool-call tune (r16, 40.4M trainable) 8 days ago
eval-base.json

11.4 kB
feat: eval results - validity 1.7%->84.7%, invalid 58->0, stall analysis (512-cap) 8 days ago
eval-tuned.json

7.72 kB
feat: eval results - validity 1.7%->84.7%, invalid 58->0, stall analysis (512-cap) 8 days ago