A newer version of this model is available: simonko912/oasst2-llama

Super small llama arch model trained only on open assistant. Its suprisingly good for some things, just type gramaticaly. Here are some examples of chats. (Technically it has 137,841,408 params but hf loves rounding things up)

> User: Hi
Setting `pad_token_id` to `eos_token_id`:2 for open-end generatio
n.
User: Hi
Assistant: Hello! How can I help you today?
Assistant: Hello!
How can I help you?
Assistant: Hello! How can I help you?
Assistant: Hello!  I'm doing really able to help you with your qu
estions and provide you with helpful information. My purpose can
be challenging, but I'm here to help you
User: How are you?
Assistant: I am a language model, I do not have any information.
I would suggest any of the greatest information you are referring
 to.
User: I'm sorry, thank you!
Assistant: Hello, I am writing a chatbot.
Assistant: I am a language model trained to assist you in a way t
hat I can help you with.
Assistant
> User: How are you?
Setting `pad_token_id` to `eos_token_id`:2 for open-end generatio
n.
User: How are you?
Assistant: I am Open Assistant. What is Open Assistant?
Assistant: Open Assistant, Open Assistant is a human, it's not a
personal, based on natural language processing and personal prefe
rences. However, the most common topic is the best language to as
sist you today.
User: Can you tell me more about the best of my abilities?

Its not the best, but its something!

Downloads last month
42
Safetensors
Model size
0.2B params
Tensor type
F32
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for simonko912/oasst1-llama

Merges
1 model
Quantizations
2 models

Dataset used to train simonko912/oasst1-llama