v1.0

inference/vllm/vllm/transformers_utils/configs/__init__.py

+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.cpm import CPMDragonflyConfig
+from vllm.transformers_utils.configs.cpm_mistral import CPMMistralConfig
+
+# 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.mpt import MPTConfig
+from vllm.transformers_utils.configs.qwen import QWenConfig
+from vllm.transformers_utils.configs.yi import YiConfig
+
+__all__ = [
+    "AquilaConfig",
+    "BaiChuanConfig",
+    "ChatGLMConfig",
+    "CPMDragonflyConfig",
+    "CPMMistralConfig",
+    "MPTConfig",
+    "QWenConfig",
+    "RWConfig",
+    "YiConfig",
+]

inference/vllm/vllm/transformers_utils/configs/aquila.py

+# 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,
+        )

inference/vllm/vllm/transformers_utils/configs/baichuan.py

+# 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,
+        )

inference/vllm/vllm/transformers_utils/configs/chatglm.py

+# coding=utf-8
+# Adapted from
+# https://github.com/THUDM/ChatGLM2-6B
+from transformers import PretrainedConfig
+
+
+class ChatGLMConfig(PretrainedConfig):
+    model_type = "chatglm"
+    attribute_map = {
+        "num_hidden_layers": "num_layers",
+        "n_head_kv": "multi_query_group_num",
+    }
+
+    def __init__(self,
+                 num_layers=28,
+                 padded_vocab_size=65024,
+                 hidden_size=4096,
+                 ffn_hidden_size=13696,
+                 kv_channels=128,
+                 num_attention_heads=32,
+                 seq_length=2048,
+                 hidden_dropout=0.0,
+                 attention_dropout=0.0,
+                 layernorm_epsilon=1e-5,
+                 rmsnorm=True,
+                 apply_residual_connection_post_layernorm=False,
+                 post_layer_norm=True,
+                 add_bias_linear=False,
+                 add_qkv_bias=False,
+                 interleaved_qkv=False,
+                 bias_dropout_fusion=True,
+                 multi_query_attention=False,
+                 multi_query_group_num=1,
+                 apply_query_key_layer_scaling=True,
+                 attention_softmax_in_fp32=True,
+                 fp32_residual_connection=False,
+                 quantization_bit=0,
+                 pre_seq_len=None,
+                 prefix_projection=False,
+                 **kwargs):
+        self.num_layers = num_layers
+        self.vocab_size = padded_vocab_size
+        self.padded_vocab_size = padded_vocab_size
+        self.hidden_size = hidden_size
+        self.ffn_hidden_size = ffn_hidden_size
+        self.kv_channels = kv_channels
+        self.num_attention_heads = num_attention_heads
+        self.seq_length = seq_length
+        self.hidden_dropout = hidden_dropout
+        self.attention_dropout = attention_dropout
+        self.layernorm_epsilon = layernorm_epsilon
+        self.rmsnorm = rmsnorm
+        self.apply_residual_connection_post_layernorm = (
+            apply_residual_connection_post_layernorm)
+        self.post_layer_norm = post_layer_norm
+        self.add_bias_linear = add_bias_linear
+        self.add_qkv_bias = add_qkv_bias
+        self.bias_dropout_fusion = bias_dropout_fusion
+        self.multi_query_attention = multi_query_attention
+        self.multi_query_group_num = multi_query_group_num
+        self.apply_query_key_layer_scaling = apply_query_key_layer_scaling
+        self.attention_softmax_in_fp32 = attention_softmax_in_fp32
+        self.fp32_residual_connection = fp32_residual_connection
+        self.quantization_bit = quantization_bit
+        self.pre_seq_len = pre_seq_len
+        self.prefix_projection = prefix_projection
+        self.interleaved_qkv = interleaved_qkv
+        super().__init__(**kwargs)

inference/vllm/vllm/transformers_utils/configs/cpm.py

+# coding=utf-8
+# Copyright 2022 The OpenBMB team.
+#
+# 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 typing import List, Optional, Tuple
+
+import torch
+import torch.nn.functional as F
+from transformers.configuration_utils import PretrainedConfig
+from typing_extensions import TypedDict
+
+
+class CPMDragonflyConfig(PretrainedConfig):
+    model_type = "cpm_dragonfly"
+    keys_to_ignore_at_inference = ["past_key_values"]
+    attribute_map = {
+        "num_key_value_heads": "num_kv_heads",
+        "hidden_act": "activate_fn",
+        "hidden_size": "dim_model",
+        "num_attention_heads": "num_heads",
+        "intermediate_size": "dim_ff",
+        "num_hidden_layers": "num_layers",
+        "vocab_size": "vocab_size",
+        "rms_norm_eps": "eps",
+        "scale_emb": "scale_emb",
+        "scale_depth": "scale_depth",
+        "scale": "scale",
+        "attention_scale": "attention_scale",
+        "qk_norm": "qk_norm",
+        "ffn_gated": "ffn_gated",
+    } # model specific to common
+
+    def __init__(
+        self,
+        vocab_size=32000,
+        dim_model=4096,
+        num_heads=32,
+        num_kv_heads=32,
+        dim_head=128,
+        dim_ff=11008,
+        num_layers=32,
+        dropout_p=0.0,
+        activate_fn="silu",
+        scale=True,
+        scale_emb: float=1.,
+        scale_depth: float=-1,
+        dim_model_base:int=None,
+        eps=1e-5,
+        init_std=0.02,
+        half: bool = True,
+        half_type = 'bf16',
+        mask_modules: Optional[List[Tuple[bool, bool]]] = None,
+        use_flash_attn: bool = True,
+        flash_attn_mask_shape="1d",
+        flash_impl="cuda",
+        base=10000,
+        non_checkpointing_layers_num:int = 0,
+        attention_scale=1,
+        qk_norm=False,
+        ffn_gated=True,
+        tie_lm_head=False,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.dim_model = dim_model
+        self.num_heads = num_heads
+        self.num_kv_heads = num_kv_heads
+        self.dim_head = dim_head
+        self.dim_ff = dim_ff
+        self.num_layers = num_layers
+        self.dropout_p = dropout_p
+        self.activate_fn = activate_fn
+        self.scale = scale
+        self.scale_emb = scale_emb
+        self.half = half
+        self.half_type = half_type
+        self.dim_model_base = dim_model_base
+        self.scale_depth = scale_depth
+        self.eps = eps
+        self.init_std = init_std
+        self.flash_impl = flash_impl
+        self.mask_modules = mask_modules
+        self.use_flash_attn = use_flash_attn
+        self.flash_attn_mask_shape = flash_attn_mask_shape
+        self.base = base
+        self.attention_scale=attention_scale
+        self.qk_norm = qk_norm
+        self.ffn_gated = ffn_gated
+        self.non_checkpointing_layers_num = non_checkpointing_layers_num
+        self.tie_lm_head = tie_lm_head
+        self.use_bfloat16 = True if self.half_type == 'bf16' else False
+        print("gated or not {}, tie or not {}, qk_norm {}".format(self.ffn_gated, self.tie_lm_head, self.qk_norm))
+        super().__init__(architectures=["CPMDragonflyForCausalLM"])
+    
+    @property
+    def scale_width(self,):
+        if self.scale:
+            return self.dim_model / self.dim_model_base
+        else:
+            return 1.
+    
+    @property
+    def dtype(self, ):
+        if self.half:
+            if self.half_type == 'bf16':
+                return torch.bfloat16
+            else:
+                return torch.half
+        else:
+            return torch.float
+    
\ No newline at end of file

inference/vllm/vllm/transformers_utils/configs/cpm_mistral.py

+# coding=utf-8
+# Copyright 2022 The OpenBMB team.
+#
+# 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 typing import List, Optional, Tuple
+
+import torch
+import torch.nn.functional as F
+from transformers.configuration_utils import PretrainedConfig
+from typing_extensions import TypedDict
+
+
+class CPMMistralConfig(PretrainedConfig):
+    model_type = "cpm_mistral"
+    keys_to_ignore_at_inference = ["past_key_values"]
+    attribute_map = {
+        "num_key_value_heads": "num_kv_heads",
+        "hidden_act": "activate_fn",
+        "hidden_size": "dim_model",
+        "num_attention_heads": "num_heads",
+        "intermediate_size": "dim_ff",
+        "num_hidden_layers": "num_layers",
+        "vocab_size": "vocab_size",
+        "rms_norm_eps": "eps",
+        "scale_emb": "scale_emb",
+        "scale_depth": "scale_depth",
+        "scale": "scale",
+        "attention_scale": "attention_scale"
+    }
+
+    def __init__(
+        self,
+        vocab_size=32000,
+        dim_model=4096,
+        num_heads=32,
+        num_kv_heads=32,
+        dim_head=128,
+        dim_ff=11008,
+        num_layers=32,
+        dropout_p=0.0,
+        activate_fn="silu",
+        scale=True,
+        scale_emb: float=1.,
+        scale_depth: float=-1,
+        dim_model_base:int=None,
+        eps=1e-5,
+        init_std=0.02,
+        half: bool = True,
+        half_type = 'bf16',
+        mask_modules: Optional[List[Tuple[bool, bool]]] = None,
+        use_flash_attn: bool = True,
+        flash_attn_mask_shape="1d",
+        flash_impl="cuda",
+        base=10000,
+        non_checkpointing_layers_num:int = 0,
+        attention_scale=1,
+        max_position_embeddings=8192,
+        rope_scaling=None,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.dim_model = dim_model
+        self.num_heads = num_heads
+        self.num_kv_heads = num_kv_heads
+        self.dim_head = dim_head
+        self.dim_ff = dim_ff
+        self.num_layers = num_layers
+        self.dropout_p = dropout_p
+        self.activate_fn = activate_fn
+        self.scale = scale
+        self.scale_emb = scale_emb
+        self.half = half
+        self.half_type = half_type
+        self.dim_model_base = dim_model_base
+        self.scale_depth = scale_depth
+        self.eps = eps
+        self.init_std = init_std
+        self.flash_impl = flash_impl
+        self.mask_modules = mask_modules
+        self.use_flash_attn = use_flash_attn
+        self.flash_attn_mask_shape = flash_attn_mask_shape
+        self.base = base
+        self.attention_scale=attention_scale
+        self.max_position_embeddings = max_position_embeddings
+        self.non_checkpointing_layers_num = non_checkpointing_layers_num
+        self.rope_scaling = rope_scaling
+        super().__init__(architectures=["CPMMistralForCausalLM"])
+    
+    @property
+    def scale_width(self,):
+        if self.scale:
+            return self.dim_model / self.dim_model_base
+        else:
+            return 1.
+    
+    @property
+    def dtype(self, ):
+        if self.half:
+            if self.half_type == 'bf16':
+                return torch.bfloat16
+            else:
+                return torch.half
+        else:
+            return torch.float
+    
\ No newline at end of file

inference/vllm/vllm/transformers_utils/configs/cpmmistral.py

+# coding=utf-8
+# Copyright 2022 The OpenBMB team.
+#
+# 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 typing import List
+from typing import Optional
+from typing import Tuple
+
+import torch
+import torch.nn.functional as F
+from typing_extensions import TypedDict
+from transformers.configuration_utils import PretrainedConfig
+
+
+class CPMMistralConfig(PretrainedConfig):
+    model_type = "cpmmistral"
+    keys_to_ignore_at_inference = ["past_key_values"]
+    attribute_map = {
+        "num_key_value_heads": "num_kv_heads",
+        "hidden_act": "activate_fn",
+        "hidden_size": "dim_model",
+        "num_attention_heads": "num_heads",
+        "intermediate_size": "dim_ff",
+        "num_hidden_layers": "num_layers",
+        "vocab_size": "vocab_size",
+        "rms_norm_eps": "eps",
+        "scale_emb": "scale_emb",
+        "scale_depth": "scale_depth",
+        "scale": "scale",
+        "attention_scale": "attention_scale"
+    }
+
+    def __init__(
+        self,
+        vocab_size=32000,
+        dim_model=4096,
+        num_heads=32,
+        num_kv_heads=32,
+        dim_head=128,
+        dim_ff=11008,
+        num_layers=32,
+        dropout_p=0.0,
+        activate_fn="silu",
+        scale=True,
+        scale_emb: float=1.,
+        scale_depth: float=-1,
+        dim_model_base:int=None,
+        eps=1e-5,
+        init_std=0.02,
+        half: bool = True,
+        half_type = 'bf16',
+        mask_modules: Optional[List[Tuple[bool, bool]]] = None,
+        use_flash_attn: bool = True,
+        flash_attn_mask_shape="1d",
+        flash_impl="cuda",
+        base=10000,
+        non_checkpointing_layers_num:int = 0,
+        attention_scale=1,
+        max_position_embeddings=8192,
+        rope_scaling=None,
+        **kwargs,
+    ):
+        self.vocab_size = vocab_size
+        self.dim_model = dim_model
+        self.num_heads = num_heads
+        self.num_kv_heads = num_kv_heads
+        self.dim_head = dim_head
+        self.dim_ff = dim_ff
+        self.num_layers = num_layers
+        self.dropout_p = dropout_p
+        self.activate_fn = activate_fn
+        self.scale = scale
+        self.scale_emb = scale_emb
+        self.half = half
+        self.half_type = half_type
+        self.dim_model_base = dim_model_base
+        self.scale_depth = scale_depth
+        self.eps = eps
+        self.init_std = init_std
+        self.flash_impl = flash_impl
+        self.mask_modules = mask_modules
+        self.use_flash_attn = use_flash_attn
+        self.flash_attn_mask_shape = flash_attn_mask_shape
+        self.base = base
+        self.attention_scale=attention_scale
+        self.max_position_embeddings = max_position_embeddings
+        self.non_checkpointing_layers_num = non_checkpointing_layers_num
+        self.rope_scaling = rope_scaling
+        super().__init__(architectures=["CPMMistralForCausalLM"])
+    
+    @property
+    def scale_width(self,):
+        if self.scale:
+            return self.dim_model / self.dim_model_base
+        else:
+            return 1.
+    
+    @property
+    def dtype(self, ):
+        if self.half:
+            if self.half_type == 'bf16':
+                return torch.bfloat16
+            else:
+                return torch.half
+        else:
+            return torch.float
+    
\ No newline at end of file

inference/vllm/vllm/transformers_utils/configs/falcon.py

+# Adapted from
+# https://huggingface.co/tiiuae/falcon-7b/blob/main/configuration_RW.py
+# Copyright 2023 The vLLM team.
+# Copyright 2022 the Big Science Workshop and HuggingFace Inc. team.
+# All rights reserved.
+#
+# 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.
+"""Falcon configuration"""
+from transformers.configuration_utils import PretrainedConfig
+
+
+class RWConfig(PretrainedConfig):
+    model_type = "falcon"
+    keys_to_ignore_at_inference = ["past_key_values"]
+    attribute_map = {
+        "num_hidden_layers": "n_layer",
+        "num_attention_heads": "n_head",
+        "num_kv_heads": "n_head_kv",
+    }
+
+    def __init__(
+        self,
+        vocab_size=250880,
+        hidden_size=64,
+        n_layer=2,
+        n_head=8,
+        layer_norm_epsilon=1e-5,
+        initializer_range=0.02,
+        use_cache=True,
+        bos_token_id=1,
+        eos_token_id=2,
+        hidden_dropout=0.0,
+        attention_dropout=0.0,
+        multi_query=True,
+        n_head_kv=None,
+        alibi=False,
+        bias=False,
+        parallel_attn=False,
+        new_decoder_architecture=False,
+        **kwargs,
+    ) -> None:
+        self.vocab_size = vocab_size
+        # Backward compatibility with n_embed kwarg
+        n_embed = kwargs.pop("n_embed", None)
+        self.hidden_size = hidden_size if n_embed is None else n_embed
+        self.n_layer = n_layer
+        self.n_head = n_head
+        self.layer_norm_epsilon = layer_norm_epsilon
+        self.initializer_range = initializer_range
+        self.use_cache = use_cache
+        self.hidden_dropout = hidden_dropout
+        self.attention_dropout = attention_dropout
+
+        self.bos_token_id = bos_token_id
+        self.eos_token_id = eos_token_id
+        self.multi_query = multi_query
+        self.n_head_kv = 1 if n_head_kv is None else n_head_kv
+        self.alibi = alibi
+        self.bias = bias
+        self.parallel_attn = parallel_attn
+        self.new_decoder_architecture = new_decoder_architecture
+
+        if self.hidden_size == 8192:
+            # Hack for falcon-40b
+            self.new_decoder_architecture = True
+
+        super().__init__(bos_token_id=bos_token_id,
+                         eos_token_id=eos_token_id,
+                         **kwargs)
+
+    @property
+    def head_dim(self):
+        return self.hidden_size // self.n_head
+
+    @property
+    def rotary(self):
+        return not self.alibi

inference/vllm/vllm/transformers_utils/configs/mpt.py

+# coding=utf-8
+# Copied from
+# https://huggingface.co/mosaicml/mpt-7b/blob/main/configuration_mpt.py
+"""A HuggingFace-style model configuration."""
+import warnings
+from typing import Any, Dict, Optional, Union
+from transformers import PretrainedConfig
+
+attn_config_defaults: Dict = {
+    'attn_type': 'multihead_attention',
+    'attn_pdrop': 0.0,
+    'attn_impl': 'triton',
+    'qk_ln': False,
+    'clip_qkv': None,
+    'softmax_scale': None,
+    'prefix_lm': False,
+    'attn_uses_sequence_id': False,
+    'alibi': False,
+    'alibi_bias_max': 8
+}
+ffn_config_defaults: Dict = {'ffn_type': 'mptmlp'}
+init_config_defaults: Dict = {
+    'name': 'kaiming_normal_',
+    'fan_mode': 'fan_in',
+    'init_nonlinearity': 'relu',
+    'init_div_is_residual': True,
+    'emb_init_std': None,
+    'emb_init_uniform_lim': None,
+    'init_std': None,
+    'init_gain': 0.0
+}
+
+
+class MPTConfig(PretrainedConfig):
+    model_type = 'mpt'
+    attribute_map = {
+        'num_attention_heads': 'n_heads',
+        'hidden_size': 'd_model',
+        'num_hidden_layers': 'n_layers',
+    }
+
+    # pylint: disable=dangerous-default-value
+    def __init__(self,
+                 d_model: int = 2048,
+                 n_heads: int = 16,
+                 n_layers: int = 24,
+                 expansion_ratio: int = 4,
+                 max_seq_len: int = 2048,
+                 vocab_size: int = 50368,
+                 resid_pdrop: float = 0.0,
+                 emb_pdrop: float = 0.0,
+                 learned_pos_emb: bool = True,
+                 attn_config: Dict = attn_config_defaults,
+                 ffn_config: Dict = ffn_config_defaults,
+                 init_device: str = 'cpu',
+                 logit_scale: Optional[Union[float, str]] = None,
+                 no_bias: bool = False,
+                 embedding_fraction: float = 1.0,
+                 norm_type: str = 'low_precision_layernorm',
+                 use_cache: bool = False,
+                 init_config: Dict = init_config_defaults,
+                 fc_type: str = 'torch',
+                 verbose: Optional[int] = None,
+                 **kwargs: Any):
+        # pylint: disable=line-too-long
+        """The MPT configuration class.
+        Args:
+            d_model (int): The size of the embedding dimension of the model.
+            n_heads (int): The number of attention heads.
+            n_layers (int): The number of layers in the model.
+            expansion_ratio (int): The ratio of the up/down scale in the ffn.
+            max_seq_len (int): The maximum sequence length of the model.
+            vocab_size (int): The size of the vocabulary.
+            resid_pdrop (float): The dropout probability applied to the attention output before combining with residual.
+            emb_pdrop (float): The dropout probability for the embedding layer.
+            learned_pos_emb (bool): Whether to use learned positional embeddings
+            attn_config (Dict): A dictionary used to configure the model's attention module:
+                attn_type (str): type of attention to use. Options: multihead_attention, multiquery_attention, grouped_query_attention
+                attn_pdrop (float): The dropout probability for the attention layers.
+                attn_impl (str): The attention implementation to use. One of 'torch', 'flash', or 'triton'.
+                qk_ln (bool): Whether to apply layer normalization to the queries and keys in the attention layer.
+                clip_qkv (Optional[float]): If not None, clip the queries, keys, and values in the attention layer to
+                    this value.
+                softmax_scale (Optional[float]): If not None, scale the softmax in the attention layer by this value. If None,
+                    use the default scale of ``1/sqrt(d_keys)``.
+                prefix_lm (Optional[bool]): Whether the model should operate as a Prefix LM. This requires passing an
+                    extra `prefix_mask` argument which indicates which tokens belong to the prefix. Tokens in the prefix
+                    can attend to one another bi-directionally. Tokens outside the prefix use causal attention.
+                attn_uses_sequence_id (Optional[bool]): Whether to restrict attention to tokens that have the same sequence_id.
+                    When the model is in `train` mode, this requires passing an extra `sequence_id` argument which indicates
+                    which sub-sequence each token belongs to.
+                    Defaults to ``False`` meaning any provided `sequence_id` will be ignored.
+                alibi (bool): Whether to use the alibi bias instead of position embeddings.
+                alibi_bias_max (int): The maximum value of the alibi bias.
+                kv_n_heads (Optional[int]): For grouped_query_attention only, allow user to specify number of kv heads.
+            ffn_config (Dict): A dictionary used to configure the model's ffn module:
+                ffn_type (str): type of ffn to use. Options: mptmlp, te_ln_mlp
+            init_device (str): The device to use for parameter initialization.
+            logit_scale (Optional[Union[float, str]]): If not None, scale the logits by this value.
+            no_bias (bool): Whether to use bias in all layers.
+            verbose (int): The verbosity level. 0 is silent.
+            embedding_fraction (float): The fraction to scale the gradients of the embedding layer by.
+            norm_type (str): choose type of norm to use
+            use_cache (bool): Whether or not the model should return the last key/values attentions
+            init_config (Dict): A dictionary used to configure the model initialization:
+                init_config.name: The parameter initialization scheme to use. Options: 'default_', 'baseline_',
+                    'kaiming_uniform_', 'kaiming_normal_', 'neox_init_', 'small_init_', 'xavier_uniform_', or
+                    'xavier_normal_'. These mimic the parameter initialization methods in PyTorch.
+                init_div_is_residual (Union[int, float, str, bool]): Value to divide initial weights by if ``module._is_residual`` is True.
+                emb_init_std (Optional[float]): The standard deviation of the normal distribution used to initialize the embedding layer.
+                emb_init_uniform_lim (Optional[Union[Tuple[float, float], float]]): The lower and upper limits of the uniform distribution
+                    used to initialize the embedding layer. Mutually exclusive with ``emb_init_std``.
+                init_std (float): The standard deviation of the normal distribution used to initialize the model,
+                    if using the baseline_ parameter initialization scheme.
+                init_gain (float): The gain to use for parameter initialization with kaiming or xavier initialization schemes.
+                fan_mode (str): The fan mode to use for parameter initialization with kaiming initialization schemes.
+                init_nonlinearity (str): The nonlinearity to use for parameter initialization with kaiming initialization schemes.
+                ---
+                See llmfoundry.models.utils.param_init_fns.py for info on other param init config options
+            fc_type (str): choose fc layer implementation. Options: torch and te. te layers support fp8 when using H100 GPUs.
+        """
+        self.d_model = d_model
+        self.n_heads = n_heads
+        self.n_layers = n_layers
+        self.expansion_ratio = expansion_ratio
+        self.max_seq_len = max_seq_len
+        self.vocab_size = vocab_size
+        self.resid_pdrop = resid_pdrop
+        self.emb_pdrop = emb_pdrop
+        self.learned_pos_emb = learned_pos_emb
+        self.attn_config = attn_config
+        self.ffn_config = ffn_config
+        self.init_device = init_device
+        self.logit_scale = logit_scale
+        self.no_bias = no_bias
+        self.embedding_fraction = embedding_fraction
+        self.norm_type = norm_type
+        self.use_cache = use_cache
+        self.init_config = init_config
+        self.fc_type = fc_type
+        if verbose is not None:
+            warnings.warn(
+                DeprecationWarning(
+                    'verbose argument for MPTConfig is now ignored and will be removed. Use python_log_level instead.'
+                ))
+        if 'name' in kwargs:
+            del kwargs['name']
+        if 'loss_fn' in kwargs:
+            del kwargs['loss_fn']
+        if self.attn_config.get('alibi', False):
+            self.learned_pos_emb = False
+            warnings.warn(
+                f'alibi is turned on, setting `learned_pos_emb` to {self.learned_pos_emb}`'
+            )
+        super().__init__(**kwargs)
+        self._validate_config()
+
+    def _set_config_defaults(
+            self, config: Dict[str, Any],
+            config_defaults: Dict[str, Any]) -> Dict[str, Any]:
+        for (k, v) in config_defaults.items():
+            if k not in config:
+                config[k] = v
+        return config
+
+    def _validate_config(self) -> None:
+        self.attn_config = self._set_config_defaults(self.attn_config,
+                                                     attn_config_defaults)
+        self.ffn_config = self._set_config_defaults(self.ffn_config,
+                                                    ffn_config_defaults)
+        self.init_config = self._set_config_defaults(self.init_config,
+                                                     init_config_defaults)
+        if self.d_model % self.n_heads != 0:
+            raise ValueError('d_model must be divisible by n_heads')
+        if any((
+                prob < 0 or prob > 1 for prob in
+            [self.attn_config['attn_pdrop'], self.resid_pdrop, self.emb_pdrop]
+        )):
+            raise ValueError(
+                "self.attn_config['attn_pdrop'], resid_pdrop, emb_pdrop are probabilities and must be between 0 and 1"  # pylint: disable=line-too-long
+            )
+        if self.attn_config['attn_impl'] not in ['torch', 'flash', 'triton']:
+            raise ValueError(
+                f"Unknown attn_impl={self.attn_config['attn_impl']}")
+        if self.attn_config['prefix_lm'] and self.attn_config[
+                'attn_impl'] not in ['torch', 'triton']:
+            raise NotImplementedError(
+                'prefix_lm only implemented with torch and triton attention.')
+        if self.attn_config['alibi'] and self.attn_config['attn_impl'] not in [
+                'torch', 'triton'
+        ]:
+            raise NotImplementedError(
+                'alibi only implemented with torch and triton attention.')
+        if self.attn_config['attn_uses_sequence_id'] and self.attn_config[
+                'attn_impl'] not in ['torch', 'triton']:
+            raise NotImplementedError(
+                'attn_uses_sequence_id only implemented with torch and triton attention.'  # pylint: disable=line-too-long
+            )
+        if self.embedding_fraction > 1 or self.embedding_fraction <= 0:
+            raise ValueError(
+                'model.embedding_fraction must be between 0 (exclusive) and 1 (inclusive)!'  # pylint: disable=line-too-long
+            )
+        if isinstance(self.logit_scale,
+                      str) and self.logit_scale != 'inv_sqrt_d_model':
+            raise ValueError(
+                f"self.logit_scale={self.logit_scale!r} is not recognized as an option; use numeric value or 'inv_sqrt_d_model'."  # pylint: disable=line-too-long
+            )
+        if self.init_config.get('name', None) is None:
+            raise ValueError(
+                f"self.init_config={self.init_config!r} 'name' needs to be set."
+            )
+        if not self.learned_pos_emb and (not self.attn_config['alibi']):
+            warnings.warn(
+                'Positional information not being provided to the model.')
+        if self.fc_type == 'te' or self.ffn_config['ffn_type'] == 'te_ln_mlp':
+            try:
+                # pylint: disable=import-outside-toplevel
+                import transformer_engine.pytorch as te
+                del te
+            except Exception as exc:
+                raise ImportError(
+                    # pylint: disable=line-too-long
+                    'TransformerEngine import fail. `fc_type: te` requires TransformerEngine be installed. '
+                    +
+                    'The required version of transformer_engine also requires FlashAttention v1.0.6 is installed:\n'
+                    + 'pip install flash-attn==1.0.6 --no-build-isolation \n' +
+                    'pip install git+https://github.com/NVIDIA/TransformerEngine.git@144e4888b2cdd60bd52e706d5b7a79cb9c1a7156'
+                ) from exc
+        if self.ffn_config['ffn_type'] == 'mptmlp':
+            self.ffn_config['fc_type'] = self.fc_type
+        elif self.ffn_config['ffn_type'] == 'te_ln_mlp':
+            self.ffn_config['bias'] = not self.no_bias

inference/vllm/vllm/transformers_utils/configs/qwen.py

+# 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)