CoolFace
Modelpublic

Flexan/HumanPet-X1-1.7B

sourceHugging Facecc-by-sa-4.0updated 5mo agoView on Hugging Face
0likes19downloads
Model Card

HumanPet X1 1.7B

Description

HumanPet X1 1.7B is an instruct LLM consisting of 1.7B parameters trained to talk in a human conversational manner. It does not support reasoning nor tool-calling (although the base model does). The model was LoRA fine-tuned with Qwen/Qwen3-1.7B as base model.

The HumanPet series is part of an experiment. Do not expect consistent releases.

Explanation of the experiment Progress on the experiment Findings for HumanPet X1 1.7B

Chat Format

HumanPet X1 1.7B uses the ChatML format, e.g.:

text
<|im_start|>system
System message<|im_end|>
<|im_start|>user
User prompt<|im_end|>
<|im_start|>assistant
Assistant response<|im_end|>

Usage

This model is not trained on system prompts. Therefore, it is recommended to not send any system messages. This includes tools (which this model also was not trained on).

The assistant response has the following format:

text
<|im_start|>assistant
<think>

</think>

What happened?

<extra>
{"intent": "question", "emotion": "no emotion"}
</extra><|im_end|>

Note that the <think>...</think> tags are always empty, as this model was not trained on reasoning data. The <extra>...</extra> tags contain JSON data of this schema:

ts
{
   intent: "directive" | "commissive" | "question" | "inform",
   emotion: "no emotion" | "happiness" | "surprise" | "fear" | "disgust" | "sadness" | "anger"
}

If the model does not adhere to this schema, please let us know in the community tab.

Datasets

  1. 1.[ConvLab/dailydialog](https://huggingface.co/datasets/ConvLab/dailydialog) 4.3k chats Only the conversations from the "Relationship" domain have been used.