lora-slm-training
github.com/saurabhzx/lora-slm-training
d3cb3dcmastersynced 1w ago
Code HealthView report
Average health
9.6/10
Excellent
Hotspot health
10.0/10
Excellent
CostsView costs →
What this repo's indexing and doc generation cost, per sync.
Open the Costs page →Attention Needed9
scripts/generate_prod_logs.pyKnowledge Silo
100% single-owner concentration
scripts/generate_training_data.pyKnowledge Silo
100% single-owner concentration
scripts/get_cleaned_training_data.pyKnowledge Silo
100% single-owner concentration
scripts/inference_test_tiny_llma_fixed.pyKnowledge Silo
100% single-owner concentration
scripts/train_lora.pyKnowledge Silo
100% single-owner concentration
No commit history
Commit activity appears once git history is indexed.
Recent DecisionsView all
Use synthetic production-like logs for training dataproposed
1w ago · readme miningTarget CPU and consumer hardware (including Apple MPS)proposed
1w ago · readme miningMerge LoRA into base model for standalone inferenceproposed
1w ago · readme miningUse HuggingFace Transformers, PEFT, and Accelerateproposed
1w ago · readme miningUse LoRA for parameter-efficient fine-tuningproposed
1w ago · readme miningNo hotspot data
No churn or complexity hotspots detected in this repository.