|
"""Mistral Model.""" |
|
|
|
import warnings |
|
|
|
from megatron import get_args |
|
|
|
from . import GPTModel |
|
from .enums import PositionEmbeddingType |
|
|
|
|
|
class MistralModel(GPTModel): |
|
def __init__( |
|
self, |
|
num_tokentypes: int = 0, |
|
parallel_output: bool = True, |
|
pre_process: bool = True, |
|
post_process: bool = True, |
|
model_type=None, |
|
): |
|
args = get_args() |
|
|
|
|
|
assert ( |
|
args.position_embedding_type == PositionEmbeddingType.rotary |
|
), f"Mistral uses rotary embedding, not {args.position_embedding_type}" |
|
assert not args.use_post_ln, "Mistral does not use post_ln" |
|
assert args.glu_activation == "swiglu", "Mistral works with swiglu activation" |
|
assert not args.use_bias, "Mistral does not use bias" |
|
assert not args.parallel_attn, "Mistral does not use parallel_attn" |
|
assert args.use_rms_norm, "Mistral uses rms_norm" |
|
assert not args.tie_embed_logits, "Mistral unties embedding and lm_head weights" |
|
assert ( |
|
args.sliding_window_size == 4096 |
|
), "Mistral uses sliding window attention (sliding_window=4096)" |
|
|
|
|
|
if not args.use_flash_attn: |
|
warnings.warn( |
|
"Mistral should use flash attn (for sliding window local attention)" |
|
) |
|
|
|
if args.bias_gelu_fusion: |
|
warnings.warn("Mistral is not intended to use bias_gelu_fusion") |
|
if args.bias_dropout_fusion: |
|
warnings.warn("Mistral is not intended to use bias_dropout_fusion") |
|
if args.hidden_dropout > 0.0 and not args.lima_dropout: |
|
warnings.warn("Mistral is not intended to use dropout") |
|
if args.attention_dropout > 0.0: |
|
warnings.warn("Mistral is not intended to use dropout") |
|
super().__init__( |
|
num_tokentypes=num_tokentypes, |
|
parallel_output=parallel_output, |
|
pre_process=pre_process, |
|
post_process=post_process, |
|
model_type=model_type, |
|
) |
|
|
