Refactor of models and trainers with base class for common methods

[PierpaoloSorbellini]

• Refactor models and trainers to avoid code replication.
• Added logs with loguru package.
• Fix logs with MultiGPU trainers.
• Added support for LoRA with PEFT library.
• Added support for load_8bit option with HF models.
• Added self-instruct dataset of HF.
• Added CerebrasGPT and Decapoda LLaMA models from HF.
• Added mixed-precision training to reduce GPU memory requirements.
• Fixed RLHF KL divergence equation.
• Added support to keep only the last n checkpoints for all training.
• Added generation of negative examples when creating the reward dataset to improve the quality of the reward model.
• Improved stability of MultiGPU training with both Accelerate form HF and DeepSpeed.

[diegofiori]

LGTM!

[diegofiori]

do we need to auto-cast to fp16 all the tensors? Shouldn't this be a config param?

[PierpaoloSorbellini]

just following documentation...
https://pytorch.org/docs/stable/notes/amp_examples.html
wrt to casting manually the tensors, this is better with less problem with types in the embedding.
It is not a config param because if you do not use fp16 you would use fp32 and is probably worse.
not seen the point of adding the option for fp32.

[diegofiori]

But what if I want to train the model in fp32 precision? (DeepSpeed for instance allows the user to select the precision)

[diegofiori]

@PierpaoloSorbellini please add a description of what this PR is adding in terms of features and which bugs it is fixing.