Poddle-2B-v1

ToasterAI debut!

Toaster

A Gemma 2B base fine-tune designed to replicate the style and prose of Dopple.AI. It's not really meant for proper usage, and it's still rather unstable. There wasn't a lot of data, so the resulting model is somewhat distinct from Dopple. It was also deliberately trained to sometimes avoid NSFW conversation to enhance the "dopple-ness" of this model. Enjoy!

This model has been fine-tuned using Unsloth with a rank and alpha of 32 for 2 epochs.

Format - ChatML:

<|im_start|>system
You are participating in a roleplay with User. Write detailed, natural, realistic responses.
Your character is [CHARACTER_NAME].<|im_end|>
<|im_start|>assistant
Hello!<|im_end|>
<|im_start|>user
How are you today?<|im_end|>
<|im_start|>assistant
Downloads last month
31
GGUF
Model size
2.51B params
Architecture
gemma

8-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 pipeline_tag.

Model tree for toasterai/Poddle-2B-v1-GGUF

Base model

google/gemma-2b
Quantized
(33)
this model