Initial Test version of my ColdBrew series of finetunes using Llama3.2 3B This version was trained for 260 steps over a Reflection dataset but hasn't yet been trained on ColdBrew's usual dataset.

Currently only releasing the LoRa for testing.

Downloads last month
3
GGUF
Model size
24.3M params
Architecture
llama

16-bit

Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.

Model tree for Theros/L3.2-3B-ColdBrew-Reflect-LoRa

Quantized
(3)
this model