File size: 1,157 Bytes
0ca529a | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 | from dataclasses import dataclass, field
from typing import Optional
@dataclass
class ModelArguments:
""" Arguments for model."""
student_name_or_path: str = field(
default=None,
metadata={"help": "Student model name or path."}
)
teacher_name_or_path: str = field(
default=None,
metadata={"help": "Teacher model name or path."}
)
attn_implementation: str = field(
default="flash_attention_2",
metadata={"help": "Attention implementation (e.g., eager, flash_attention_2, flash_attention_3, kernels-community/vllm-flash-attn3)."}
)
use_liger_kernel: bool = field(
default=False,
metadata={"help": "Use Liger Kernel in LLM."}
)
lora_rank: int = field(
default=0
)
lora_alpha: int = field(
default=16
)
target_modules: str = field(
default="all-linear",
)
lora_dropout: float = field(
default=0.0,
)
disable_fast_tokenizer: bool = field(
default=False
)
enable_thinking: bool = field(
default=False,
)
ring_attn_size: int = field(
default=1
) |