image-match/python/py/Lib/site-packages/transformers/models/dots1/configuration_dots1.py

#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
#           This file was automatically generated from src/transformers/models/dots1/modular_dots1.py.
#               Do NOT edit this file manually as any edits will be overwritten by the generation of
#             the file from the modular. If any change should be done, please apply the change to the
#                          modular_dots1.py file directly. One of our CI enforces this.
#                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
# Copyright 2025 The rednote-hilab team and the 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.
from huggingface_hub.dataclasses import strict

from ...configuration_utils import PreTrainedConfig
from ...modeling_rope_utils import RopeParameters
from ...utils import auto_docstring


@auto_docstring(checkpoint="rednote-hilab/dots.llm1.base")
@strict
class Dots1Config(PreTrainedConfig):
    r"""
    n_group (`int`, *optional*, defaults to 1):
        Number of groups for routed experts.
    first_k_dense_replace (`int`, *optional*, defaults to 0):
        Number of dense layers at the beginning of the model before the first MoE layer.

    Examples:

    ```python
    >>> from transformers import Dots1Model, Dots1Config
    >>> # Initializing a Dots1 style configuration
    >>> configuration = Dots1Config()
    >>> # Accessing the model configuration
    >>> configuration = model.config
    ```
    """

    model_type = "dots1"
    keys_to_ignore_at_inference = ["past_key_values"]

    base_model_tp_plan = {
        "layers.*.self_attn.q_proj": "colwise",
        "layers.*.self_attn.k_proj": "colwise",
        "layers.*.self_attn.v_proj": "colwise",
        "layers.*.self_attn.o_proj": "rowwise",
        "layers.*.self_attn.q_norm": "replicated_with_grad_allreduce",
        "layers.*.self_attn.k_norm": "replicated_with_grad_allreduce",
        "layers.*.mlp.experts.gate_up_proj": "packed_colwise",
        "layers.*.mlp.experts.down_proj": "rowwise",
        "layers.*.mlp.experts": "moe_tp_experts",
        "layers.*.mlp.shared_experts.gate_proj": "colwise",
        "layers.*.mlp.shared_experts.up_proj": "colwise",
        "layers.*.mlp.shared_experts.down_proj": "rowwise",
        "layers.*.mlp.gate_proj": "colwise",
        "layers.*.mlp.up_proj": "colwise",
        "layers.*.mlp.down_proj": "rowwise",
    }

    base_model_pp_plan = {
        "embed_tokens": (["input_ids"], ["inputs_embeds"]),
        "layers": (["hidden_states", "attention_mask"], ["hidden_states"]),
        "norm": (["hidden_states"], ["hidden_states"]),
    }
    attribute_map = {
        "num_local_experts": "n_routed_experts",
    }

    vocab_size: int = 152064
    hidden_size: int = 4608
    intermediate_size: int = 10944
    moe_intermediate_size: int = 1408
    num_hidden_layers: int = 62
    num_attention_heads: int = 32
    num_key_value_heads: int | None = 32
    n_shared_experts: int | None = None
    n_routed_experts: int | None = None
    n_group: int | None = 1
    topk_group: int | None = 1
    num_experts_per_tok: int | None = None
    first_k_dense_replace: int | None = 0
    norm_topk_prob: bool | None = False
    hidden_act: str = "silu"
    max_position_embeddings: int = 2048
    initializer_range: float = 0.02
    rms_norm_eps: float = 1e-6
    use_cache: bool = True
    tie_word_embeddings: bool = False
    rope_parameters: RopeParameters | dict | None = None
    attention_bias: bool = False
    attention_dropout: float | int | None = 0.0
    routed_scaling_factor: float = 1.0
    sliding_window: int | None = 4096
    max_window_layers: int | None = 62
    layer_types: list[str] | None = None
    pad_token_id: int | None = None
    bos_token_id: int | None = None
    eos_token_id: int | list[int] | None = None

    def __post_init__(self, **kwargs):
        if self.num_key_value_heads is None:
            self.num_key_value_heads = self.num_attention_heads

        if self.layer_types is None:
            self.layer_types = [
                "sliding_attention"
                if self.sliding_window is not None and i >= self.max_window_layers
                else "full_attention"
                for i in range(self.num_hidden_layers)
            ]

        super().__post_init__(**kwargs)


__all__ = ["Dots1Config"]