Text Generation
Transformers
PyTorch
llama
Inference Endpoints
text-generation-inference
Edit model card

CodeLlama-7b Instruct finetuned on 6320 function calling and generic chat examples

Fine-tuned with LoRA on a small fraction of the glaive-function-calling-v2 dataset and a formatted (and slightly cleaned) version of sharegpt-hyperfiltered-3k

Prompt example:

[INST] <<SYS>>
<function>Available functions:
<function>{
    "name": "generate_password",
    "description": "Generate a random password with specified criteria",
    "parameters": {
        "type": "object",
        "properties": {
            "length": {
                "type": "integer",
                "description": "The length of the password"
            },
            "include_numbers": {
                "type": "boolean",
                "description": "Include numbers in the password"
            },
            "include_special_characters": {
                "type": "boolean",
                "description": "Include special characters in the password"
            }
        },
        "required": [
            "length"
        ]
    }
}
<</SYS>>

I need a new password. Can you generate one for me? [/INST]

The model then generates (note the leading space):

 Of course! How long would you like your password to be? And would you like it to include numbers and special characters?

If you extend the prompt to be:

[INST] <<SYS>>
<function>Available functions:
<function>{
    "name": "generate_password",
    "description": "Generate a random password with specified criteria",
    "parameters": {
        "type": "object",
        "properties": {
            "length": {
                "type": "integer",
                "description": "The length of the password"
            },
            "include_numbers": {
                "type": "boolean",
                "description": "Include numbers in the password"
            },
            "include_special_characters": {
                "type": "boolean",
                "description": "Include special characters in the password"
            }
        },
        "required": [
            "length"
        ]
    }
}
<</SYS>>

I need a new password. Can you generate one for me? [/INST] Of course! How long would you like your password to be? And would you like it to include numbers and special characters?</s><s>[INST] I'd like it to be 12 characters long. [/INST]

The model will generate (without the leading space):

<function>generate_password
{
    "length": 12
}

It can also answer questions based on a prompt without any functions:

[INST] In one sentence, what is a large language model? [/INST] A large language model is a type of artificial intelligence model that is trained on vast amounts of text data to generate human-like language and understand natural language input. [/INST] That's correct! Large language models are trained on large datasets of text to generate human-like language and understand natural language input. They are often used in applications such as chatbots, language translation, and text summarization.

A quick manual test shows that it's still able to follow a system prompt provided alongside any functions provided, including in multi-turn conversations. None of this was tested comprehensively, though, your results may vary.

Downloads last month
46
Inference API
Model is too large to load in Inference API (serverless). To try the model, launch it on Inference Endpoints (dedicated) instead.

Datasets used to train rizerphe/CodeLlama-function-calling-6320-7b-Instruct-hf