96 LoRA adapters from the main study, plus 4 from the teacher robustness test (GPT-OSS-120B as teacher). https://mdsw.tr/catalog/mdsw-1
Mehmet Davut
mehmetdavut
AI & ML interests
Founder, MDSW · Small language models, LoRA fine-tuning, synthetic data distillation
Recent Activity
updated a dataset about 19 hours ago
mehmetdavut/RubyCraft-3.4-Instruct updated a collection 12 days ago
RubyCraft-3.4-Instruct: SLM Adapters updated a collection 12 days ago
RubyCraft-3.4-Instruct: SLM AdaptersOrganizations
None yet
models 100
mehmetdavut/ruby3.4-qwen2.5-coder-1.5b-5k-hq-16bit-gpt
Updated • 3
mehmetdavut/ruby3.4-qwen2.5-coder-1.5b-5k-all-16bit-gpt
Updated • 3
mehmetdavut/ruby3.4-qwen2.5-coder-1.5b-1k-hq-16bit-gpt
Updated • 6
mehmetdavut/ruby3.4-qwen2.5-coder-1.5b-1k-all-16bit-gpt
Updated • 7
mehmetdavut/ruby3.4-qwen2.5-coder-7b-5k-hq-16bit-gemini
Updated • 3
mehmetdavut/ruby3.4-qwen2.5-coder-7b-5k-all-16bit-gemini
Updated • 3
mehmetdavut/ruby3.4-qwen2.5-7b-instruct-5k-hq-16bit-gemini
Updated • 5
mehmetdavut/ruby3.4-qwen2.5-7b-instruct-5k-all-16bit-gemini
Updated • 8
mehmetdavut/ruby3.4-llama-3.1-8b-5k-hq-16bit-gemini
Updated • 4
mehmetdavut/ruby3.4-llama-3.1-8b-5k-all-16bit-gemini
Updated • 10