feat(mistral): add Mistral model implementation and configs

- implement Mistral model in llm/models/mistral/mistral.py with GroupedQueryAttention, SwiGLU, RoPE, sliding window attention
- add __init__.py for module export
- add config files for mistral training and generation
- update universal experiment runner to support Mistral model
- add notebook for Mistral experiments
This commit is contained in:
Sergey Penkovsky
2025-10-14 14:53:45 +03:00
parent e5706a690d
commit ec0d2bd8d0
6 changed files with 3904 additions and 0 deletions

View File

@@ -42,6 +42,9 @@ def load_model_class(model_name):
elif model_name.lower() == 'llama':
from llm.models.llama import Llama
return Llama
elif model_name.lower() == 'mistral':
from llm.models.mistral import Mistral
return Mistral
else:
raise ValueError(f"Модель '{model_name}' не поддерживается.")