metadata
language: en
license: apache-2.0
tags:
- code
- qlora
- finetuned
base_model: Cartik/Sonexa-1-Chat
model_name: Sonexa 1.1 Chat
model_type: qwen3
parameters: 5B
Sonexa 1.1 Chat
This is a fine-tuned version of Sonexa-1-Chat using QLoRA on the Magicoder-Evol-Instruct-110K dataset.
Model Details
- Parameters: 5B (as specified by user)
- Training: QLoRA (4-bit quantization)
- Checkpoint: Merged v1.1