Refactor llama family models (#2637)
This commit is contained in:
@@ -5,14 +5,10 @@ from transformers import AutoConfig, PretrainedConfig
|
||||
from vllm.transformers_utils.configs import *
|
||||
|
||||
_CONFIG_REGISTRY = {
|
||||
"aquila": AquilaConfig,
|
||||
"baichuan": BaiChuanConfig,
|
||||
"chatglm": ChatGLMConfig,
|
||||
"mpt": MPTConfig,
|
||||
"qwen": QWenConfig,
|
||||
"RefinedWeb": RWConfig, # For tiiuae/falcon-40b(-instruct)
|
||||
"RefinedWebModel": RWConfig, # For tiiuae/falcon-7b(-instruct)
|
||||
"yi": YiConfig,
|
||||
}
|
||||
|
||||
|
||||
|
||||
@@ -1,20 +1,12 @@
|
||||
from vllm.transformers_utils.configs.aquila import AquilaConfig
|
||||
from vllm.transformers_utils.configs.baichuan import BaiChuanConfig
|
||||
from vllm.transformers_utils.configs.chatglm import ChatGLMConfig
|
||||
from vllm.transformers_utils.configs.mpt import MPTConfig
|
||||
from vllm.transformers_utils.configs.qwen import QWenConfig
|
||||
# RWConfig is for the original tiiuae/falcon-40b(-instruct) and
|
||||
# tiiuae/falcon-7b(-instruct) models. Newer Falcon models will use the
|
||||
# `FalconConfig` class from the official HuggingFace transformers library.
|
||||
from vllm.transformers_utils.configs.falcon import RWConfig
|
||||
from vllm.transformers_utils.configs.yi import YiConfig
|
||||
|
||||
__all__ = [
|
||||
"AquilaConfig",
|
||||
"BaiChuanConfig",
|
||||
"ChatGLMConfig",
|
||||
"MPTConfig",
|
||||
"QWenConfig",
|
||||
"RWConfig",
|
||||
"YiConfig",
|
||||
]
|
||||
|
||||
@@ -1,69 +0,0 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2023 EleutherAI and the HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
|
||||
# and OPT implementations in this library. It has been modified from its
|
||||
# original forms to accommodate minor architectural differences compared
|
||||
# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
""" Aquila model configuration"""
|
||||
|
||||
from transformers import PretrainedConfig
|
||||
|
||||
|
||||
class AquilaConfig(PretrainedConfig):
|
||||
model_type = "aquila"
|
||||
keys_to_ignore_at_inference = ["past_key_values"]
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_size=100008,
|
||||
hidden_size=4096,
|
||||
intermediate_size=11008,
|
||||
num_hidden_layers=32,
|
||||
num_attention_heads=32,
|
||||
num_key_value_heads=None,
|
||||
hidden_act="silu",
|
||||
max_position_embeddings=2048,
|
||||
initializer_range=0.006,
|
||||
rms_norm_eps=1e-5,
|
||||
use_cache=True,
|
||||
pad_token_id=0,
|
||||
bos_token_id=1,
|
||||
eos_token_id=2,
|
||||
tie_word_embeddings=False,
|
||||
**kwargs,
|
||||
):
|
||||
self.vocab_size = vocab_size
|
||||
self.max_position_embeddings = max_position_embeddings
|
||||
self.hidden_size = hidden_size
|
||||
self.intermediate_size = intermediate_size
|
||||
self.num_hidden_layers = num_hidden_layers
|
||||
# for backward compatibility
|
||||
if num_key_value_heads is None:
|
||||
num_key_value_heads = num_attention_heads
|
||||
|
||||
self.num_key_value_heads = num_key_value_heads
|
||||
self.num_attention_heads = num_attention_heads
|
||||
self.hidden_act = hidden_act
|
||||
self.initializer_range = initializer_range
|
||||
self.rms_norm_eps = rms_norm_eps
|
||||
self.use_cache = use_cache
|
||||
super().__init__(
|
||||
pad_token_id=pad_token_id,
|
||||
bos_token_id=bos_token_id,
|
||||
eos_token_id=eos_token_id,
|
||||
tie_word_embeddings=tie_word_embeddings,
|
||||
**kwargs,
|
||||
)
|
||||
@@ -1,62 +0,0 @@
|
||||
# coding=utf-8
|
||||
# Copyright 2022 EleutherAI and the HuggingFace Inc. team. All rights reserved.
|
||||
#
|
||||
# This code is based on EleutherAI's GPT-NeoX library and the GPT-NeoX
|
||||
# and OPT implementations in this library. It has been modified from its
|
||||
# original forms to accommodate minor architectural differences compared
|
||||
# to GPT-NeoX and OPT used by the Meta AI team that trained the model.
|
||||
#
|
||||
# Licensed under the Apache License, Version 2.0 (the "License");
|
||||
# you may not use this file except in compliance with the License.
|
||||
# You may obtain a copy of the License at
|
||||
#
|
||||
# http://www.apache.org/licenses/LICENSE-2.0
|
||||
#
|
||||
# Unless required by applicable law or agreed to in writing, software
|
||||
# distributed under the License is distributed on an "AS IS" BASIS,
|
||||
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
||||
# See the License for the specific language governing permissions and
|
||||
# limitations under the License.
|
||||
|
||||
from transformers.configuration_utils import PretrainedConfig
|
||||
|
||||
|
||||
class BaiChuanConfig(PretrainedConfig):
|
||||
model_type = "baichuan"
|
||||
keys_to_ignore_at_inference = ["past_key_values"]
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_size=64000,
|
||||
hidden_size=4096,
|
||||
intermediate_size=11008,
|
||||
num_hidden_layers=32,
|
||||
num_attention_heads=32,
|
||||
hidden_act="silu",
|
||||
max_position_embeddings=4096,
|
||||
initializer_range=0.02,
|
||||
rms_norm_eps=1e-6,
|
||||
use_cache=True,
|
||||
pad_token_id=0,
|
||||
bos_token_id=1,
|
||||
eos_token_id=2,
|
||||
tie_word_embeddings=False,
|
||||
**kwargs,
|
||||
):
|
||||
self.vocab_size = vocab_size
|
||||
self.max_position_embeddings = max_position_embeddings
|
||||
self.hidden_size = hidden_size
|
||||
self.intermediate_size = intermediate_size
|
||||
self.num_hidden_layers = num_hidden_layers
|
||||
self.num_attention_heads = num_attention_heads
|
||||
self.hidden_act = hidden_act
|
||||
self.initializer_range = initializer_range
|
||||
self.rms_norm_eps = rms_norm_eps
|
||||
self.use_cache = use_cache
|
||||
super().__init__(
|
||||
pad_token_id=pad_token_id,
|
||||
bos_token_id=bos_token_id,
|
||||
eos_token_id=eos_token_id,
|
||||
tie_word_embeddings=tie_word_embeddings,
|
||||
**kwargs,
|
||||
)
|
||||
@@ -1,60 +0,0 @@
|
||||
# Copyright (c) Alibaba Cloud.
|
||||
# LICENSE: https://huggingface.co/Qwen/Qwen-7B/blob/main/LICENSE
|
||||
|
||||
from transformers import PretrainedConfig
|
||||
|
||||
|
||||
class QWenConfig(PretrainedConfig):
|
||||
model_type = "qwen"
|
||||
keys_to_ignore_at_inference = ["past_key_values"]
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_size=151936,
|
||||
hidden_size=4096,
|
||||
num_hidden_layers=32,
|
||||
num_attention_heads=32,
|
||||
emb_dropout_prob=0.0,
|
||||
attn_dropout_prob=0.0,
|
||||
layer_norm_epsilon=1e-6,
|
||||
initializer_range=0.02,
|
||||
max_position_embeddings=8192,
|
||||
scale_attn_weights=True,
|
||||
use_cache=True,
|
||||
bf16=False,
|
||||
fp16=False,
|
||||
fp32=False,
|
||||
kv_channels=128,
|
||||
rotary_pct=1.0,
|
||||
rotary_emb_base=10000,
|
||||
use_dynamic_ntk=True,
|
||||
use_logn_attn=True,
|
||||
use_flash_attn="auto",
|
||||
intermediate_size=22016,
|
||||
no_bias=True,
|
||||
tie_word_embeddings=False,
|
||||
**kwargs,
|
||||
):
|
||||
self.vocab_size = vocab_size
|
||||
self.hidden_size = hidden_size
|
||||
self.intermediate_size = intermediate_size
|
||||
self.num_hidden_layers = num_hidden_layers
|
||||
self.num_attention_heads = num_attention_heads
|
||||
self.emb_dropout_prob = emb_dropout_prob
|
||||
self.attn_dropout_prob = attn_dropout_prob
|
||||
self.layer_norm_epsilon = layer_norm_epsilon
|
||||
self.initializer_range = initializer_range
|
||||
self.scale_attn_weights = scale_attn_weights
|
||||
self.use_cache = use_cache
|
||||
self.max_position_embeddings = max_position_embeddings
|
||||
self.bf16 = bf16
|
||||
self.fp16 = fp16
|
||||
self.fp32 = fp32
|
||||
self.kv_channels = kv_channels
|
||||
self.rotary_pct = rotary_pct
|
||||
self.rotary_emb_base = rotary_emb_base
|
||||
self.use_dynamic_ntk = use_dynamic_ntk
|
||||
self.use_logn_attn = use_logn_attn
|
||||
self.use_flash_attn = use_flash_attn
|
||||
self.no_bias = no_bias
|
||||
super().__init__(tie_word_embeddings=tie_word_embeddings, **kwargs)
|
||||
@@ -1,64 +0,0 @@
|
||||
""" Yi model configuration"""
|
||||
from transformers.configuration_utils import PretrainedConfig
|
||||
from transformers.utils import logging
|
||||
|
||||
logger = logging.get_logger(__name__)
|
||||
|
||||
Yi_PRETRAINED_CONFIG_ARCHIVE_MAP = {}
|
||||
|
||||
|
||||
class YiConfig(PretrainedConfig):
|
||||
r"""
|
||||
Reference:
|
||||
https://huggingface.co/01-ai/Yi-6B/blob/main/configuration_yi.py
|
||||
"""
|
||||
model_type = "Yi"
|
||||
keys_to_ignore_at_inference = ["past_key_values"]
|
||||
|
||||
def __init__(
|
||||
self,
|
||||
vocab_size=64000,
|
||||
hidden_size=4096,
|
||||
intermediate_size=11008,
|
||||
num_hidden_layers=32,
|
||||
num_attention_heads=32,
|
||||
num_key_value_heads=4,
|
||||
hidden_act="silu",
|
||||
max_position_embeddings=4096,
|
||||
initializer_range=0.02,
|
||||
rms_norm_eps=1e-5,
|
||||
use_cache=True,
|
||||
pad_token_id=0,
|
||||
bos_token_id=1,
|
||||
eos_token_id=2,
|
||||
tie_word_embeddings=False,
|
||||
output_attentions=False,
|
||||
rope_theta=5000000.0,
|
||||
**kwargs,
|
||||
):
|
||||
self.vocab_size = vocab_size
|
||||
self.max_position_embeddings = max_position_embeddings
|
||||
self.hidden_size = hidden_size
|
||||
self.intermediate_size = intermediate_size
|
||||
self.num_hidden_layers = num_hidden_layers
|
||||
self.num_attention_heads = num_attention_heads
|
||||
|
||||
# for backward compatibility
|
||||
if num_key_value_heads is None:
|
||||
num_key_value_heads = num_attention_heads
|
||||
|
||||
self.num_key_value_heads = num_key_value_heads
|
||||
self.hidden_act = hidden_act
|
||||
self.initializer_range = initializer_range
|
||||
self.rms_norm_eps = rms_norm_eps
|
||||
self.use_cache = use_cache
|
||||
self.output_attentions = output_attentions
|
||||
self.rope_theta = rope_theta
|
||||
|
||||
super().__init__(
|
||||
pad_token_id=pad_token_id,
|
||||
bos_token_id=bos_token_id,
|
||||
eos_token_id=eos_token_id,
|
||||
tie_word_embeddings=tie_word_embeddings,
|
||||
**kwargs,
|
||||
)
|
||||
Reference in New Issue
Block a user