Added int4 configs for llama3b, falcon7b and orca3b

[ljaljushkin]

What does this PR do?

Added quantization parameters for models that have the number of channels one of the Linear layer not divisible by default group size = 128.
Also fixed names of models in the DEFAULT_4BIT_CONFIGS

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests?

[ljaljushkin]

@AlexKoff88

[HuggingFaceDocBuilderDev]

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.