LFM2.5-8B-A1B-Coder
Merged model from LiquidAI/LFM2.5-8B-A1B plus:
josephmayo/LFM2.5-8B-A1B-Coder-LoRA
GGUF target:
josephmayo/LFM2.5-8B-A1B-Coder-GGUF
Training Summary
- Base model:
LiquidAI/LFM2.5-8B-A1B - Method: LoRA fine-tuning, then merged into base weights
- Training rows:
123 - Data focus: real-world and multilingual coding tasks
Validation
Validation metric:
- Base heldout mean NLL:
3.3903426826000214 - Merged model heldout mean NLL:
1.8745103478431702 - Absolute NLL decrease:
1.5158323347568512 - Relative NLL reduction:
44.71%
Public benchmark pass-rate scores are not claimed for this merged release.
Evidence files
Run evidence for this release is stored in the repository under evidence/:
These files are compact local/Kaggle run artifacts used to document training, evaluation, merge, or quantization evidence for this model family.