AFM-4.5B-Base-KDA-Only / configuration_arcee_kda.py

Duplicate from arcee-ai/AFM-4.5B-Base-KDA-Only

7c3ba06 14 days ago

9.43 kB

# 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.
"""Apriel model configuration"""

import math
from typing import Optional, Tuple

from transformers.configuration_utils import PretrainedConfig
from transformers.utils import is_torch_available, logging

logger = logging.get_logger(__name__)

if is_torch_available():
    import torch


class ArceeKDAConfig(PretrainedConfig):
    r"""
    This is the configuration class to store the configuration of a [`ArceeKDAModel`]. It is used to instantiate an Arcee
    model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
    defaults will yield a similar configuration to that of the Arcee-5B-Base.
    Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
    documentation from [`PretrainedConfig`] for more information.
    Args:
        vocab_size (`int`, *optional*, defaults to 32000):
            Vocabulary size of the Apriel model. Defines the number of different tokens that can be represented by the
            `inputs_ids` passed when calling [`AprielModel`]
        hidden_size (`int`, *optional*, defaults to 4096):
            Dimension of the hidden representations.
        intermediate_size (`int`, *optional*, defaults to 11008):
            Dimension of the MLP representations.
        num_hidden_layers (`int`, *optional*, defaults to 32):
            Number of hidden layers in the Transformer decoder.
        num_attention_heads (`int`, *optional*, defaults to 32):
            Number of attention heads for each attention layer in the Transformer decoder.
        num_key_value_heads (`int`, *optional*):
            This is the number of key_value heads that should be used to implement Grouped Query Attention. If
            `num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
            `num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
            converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
            by meanpooling all the original heads within that group. For more details checkout [this
            paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
            `num_attention_heads`.
        hidden_act (`str` or `function`, *optional*, defaults to `"silu"`):
            The non-linear activation function (function or string) in the decoder.
        max_position_embeddings (`int`, *optional*, defaults to 2048):
            The maximum sequence length that this model might ever be used with. Apriel-5B-Base supports up to 16384 tokens.
        initializer_range (`float`, *optional*, defaults to 0.02):
            The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
        rms_norm_eps (`float`, *optional*, defaults to 1e-06):
            The epsilon used by the rms normalization layers.
        use_cache (`bool`, *optional*, defaults to `True`):
            Whether or not the model should return the last key/values attentions (not used by all models). Only
            relevant if `config.is_decoder=True`.
        pad_token_id (`int`, *optional*):
            Padding token id.
        bos_token_id (`int`, *optional*, defaults to 1):
            Beginning of stream token id.
        eos_token_id (`int`, *optional*, defaults to 2):
            End of stream token id.
        pretraining_tp (`int`, *optional*, defaults to 1):
            Experimental feature. Tensor parallelism rank used during pretraining. Please refer to [this
            document](https://huggingface.co/docs/transformers/main/perf_train_gpu_many#tensor-parallelism) to
            understand more about it. This value is necessary to ensure exact reproducibility of the pretraining
            results. Please refer to [this issue](https://github.com/pytorch/pytorch/issues/76232).
        tie_word_embeddings (`bool`, *optional*, defaults to `False`):
            Whether to tie weight embeddings
        attention_bias (`bool`, *optional*, defaults to `False`):
            Whether to use a bias in the query, key, value and output projection layers during self-attention.
        attention_dropout (`float`, *optional*, defaults to 0.0):
            The dropout ratio for the attention probabilities.
        mlp_bias (`bool`, *optional*, defaults to `False`):
            Whether to use a bias in up_proj, down_proj and gate_proj layers in the MLP layers.
        head_dim (`int`, *optional*):
            The attention head dimension. If None, it will default to hidden_size // num_attention_heads
    ```python
    >>> from transformers import ArceeKDAModel, ArceeKDAConfig
    >>> # Initializing an Arcee Arcee-4B-Base style configuration
    >>> configuration = ArceeKDAConfig()
    >>> # Initializing a model from the Arcee-4B-Base style configuration
    >>> model = ArceeKDAModel(configuration)
    >>> # Accessing the model configuration
    >>> configuration = model.config
    ```"""

    model_type = "arcee_kda"
    auto_map = {
        "AutoConfig": "configuration_arcee.ArceeKDAConfig",
        "AutoModelForCausalLM": "modeling_arcee.ArceeKDAForCausalLM",
    }
    keys_to_ignore_at_inference = ["past_key_values"]
    # Default tensor parallel plan for base model `ArceeKDAModel`
    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.*.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"]),
    }

    def __init__(
        self,
        vocab_size: int = 32000,
        hidden_size: int = 2560,
        intermediate_size: int = 18432,
        num_hidden_layers: int = 32,
        num_attention_heads: int = 32,
        num_key_value_heads: Optional[int] = None,
        hidden_act: str = "relu2",
        max_position_embeddings: int = 4096,
        initializer_range: float = 0.02,
        rms_norm_eps: float = 1e-5,
        use_cache: bool = True,
        pad_token_id: Optional[int] = None,
        bos_token_id: int = 128000,
        eos_token_id: int = 128001,
        pretraining_tp: int = 1,
        tie_word_embeddings: bool = False,
        attention_bias: bool = False,
        attention_dropout: float = 0.0,
        mlp_bias: bool = False,
        head_dim: Optional[int] = None,
        linear_attn_config: Optional[dict] = None,
        **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.pretraining_tp = pretraining_tp
        self.use_cache = use_cache
        self.attention_bias = attention_bias
        self.attention_dropout = attention_dropout
        self.mlp_bias = mlp_bias
        self.head_dim = (
            head_dim
            if head_dim is not None
            else self.hidden_size // self.num_attention_heads
        )

        if linear_attn_config is not None:
            assert linear_attn_config["kda_layers"] is not None
            assert linear_attn_config["full_attn_layers"] is not None
        self.linear_attn_config = linear_attn_config

        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,
        )

    def is_kda_layer(self, layer_idx: int):
        return (
            self.linear_attn_config is not None
            and (layer_idx + 1) in self.linear_attn_config["kda_layers"]
        )


__all__ = ["ArceeKDAConfig"]

	# 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.
	"""Apriel model configuration"""

	import math
	from typing import Optional, Tuple

	from transformers.configuration_utils import PretrainedConfig
	from transformers.utils import is_torch_available, logging

	logger = logging.get_logger(__name__)

	if is_torch_available():
	import torch


	class ArceeKDAConfig(PretrainedConfig):
	r"""
	This is the configuration class to store the configuration of a [`ArceeKDAModel`]. It is used to instantiate an Arcee
	model according to the specified arguments, defining the model architecture. Instantiating a configuration with the
	defaults will yield a similar configuration to that of the Arcee-5B-Base.

	Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the
	documentation from [`PretrainedConfig`] for more information.


	Args:
	vocab_size (`int`, optional, defaults to 32000):
	Vocabulary size of the Apriel model. Defines the number of different tokens that can be represented by the
	`inputs_ids` passed when calling [`AprielModel`]
	hidden_size (`int`, optional, defaults to 4096):
	Dimension of the hidden representations.
	intermediate_size (`int`, optional, defaults to 11008):
	Dimension of the MLP representations.
	num_hidden_layers (`int`, optional, defaults to 32):
	Number of hidden layers in the Transformer decoder.
	num_attention_heads (`int`, optional, defaults to 32):
	Number of attention heads for each attention layer in the Transformer decoder.
	num_key_value_heads (`int`, optional):
	This is the number of key_value heads that should be used to implement Grouped Query Attention. If
	`num_key_value_heads=num_attention_heads`, the model will use Multi Head Attention (MHA), if
	`num_key_value_heads=1` the model will use Multi Query Attention (MQA) otherwise GQA is used. When
	converting a multi-head checkpoint to a GQA checkpoint, each group key and value head should be constructed
	by meanpooling all the original heads within that group. For more details checkout [this
	paper](https://arxiv.org/pdf/2305.13245.pdf). If it is not specified, will default to
	`num_attention_heads`.
	hidden_act (`str` or `function`, optional, defaults to `"silu"`):
	The non-linear activation function (function or string) in the decoder.
	max_position_embeddings (`int`, optional, defaults to 2048):
	The maximum sequence length that this model might ever be used with. Apriel-5B-Base supports up to 16384 tokens.
	initializer_range (`float`, optional, defaults to 0.02):
	The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
	rms_norm_eps (`float`, optional, defaults to 1e-06):
	The epsilon used by the rms normalization layers.
	use_cache (`bool`, optional, defaults to `True`):
	Whether or not the model should return the last key/values attentions (not used by all models). Only
	relevant if `config.is_decoder=True`.
	pad_token_id (`int`, optional):
	Padding token id.
	bos_token_id (`int`, optional, defaults to 1):
	Beginning of stream token id.
	eos_token_id (`int`, optional, defaults to 2):
	End of stream token id.
	pretraining_tp (`int`, optional, defaults to 1):
	Experimental feature. Tensor parallelism rank used during pretraining. Please refer to [this
	document](https://huggingface.co/docs/transformers/main/perf_train_gpu_many#tensor-parallelism) to
	understand more about it. This value is necessary to ensure exact reproducibility of the pretraining
	results. Please refer to [this issue](https://github.com/pytorch/pytorch/issues/76232).
	tie_word_embeddings (`bool`, optional, defaults to `False`):
	Whether to tie weight embeddings
	attention_bias (`bool`, optional, defaults to `False`):
	Whether to use a bias in the query, key, value and output projection layers during self-attention.
	attention_dropout (`float`, optional, defaults to 0.0):
	The dropout ratio for the attention probabilities.
	mlp_bias (`bool`, optional, defaults to `False`):
	Whether to use a bias in up_proj, down_proj and gate_proj layers in the MLP layers.
	head_dim (`int`, optional):
	The attention head dimension. If None, it will default to hidden_size // num_attention_heads

	```python
	>>> from transformers import ArceeKDAModel, ArceeKDAConfig

	>>> # Initializing an Arcee Arcee-4B-Base style configuration
	>>> configuration = ArceeKDAConfig()

	>>> # Initializing a model from the Arcee-4B-Base style configuration
	>>> model = ArceeKDAModel(configuration)

	>>> # Accessing the model configuration
	>>> configuration = model.config
	```"""

	model_type = "arcee_kda"
	auto_map = {
	"AutoConfig": "configuration_arcee.ArceeKDAConfig",
	"AutoModelForCausalLM": "modeling_arcee.ArceeKDAForCausalLM",
	}
	keys_to_ignore_at_inference = ["past_key_values"]
	# Default tensor parallel plan for base model `ArceeKDAModel`
	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.*.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"]),
	}

	def __init__(
	self,
	vocab_size: int = 32000,
	hidden_size: int = 2560,
	intermediate_size: int = 18432,
	num_hidden_layers: int = 32,
	num_attention_heads: int = 32,
	num_key_value_heads: Optional[int] = None,
	hidden_act: str = "relu2",
	max_position_embeddings: int = 4096,
	initializer_range: float = 0.02,
	rms_norm_eps: float = 1e-5,
	use_cache: bool = True,
	pad_token_id: Optional[int] = None,
	bos_token_id: int = 128000,
	eos_token_id: int = 128001,
	pretraining_tp: int = 1,
	tie_word_embeddings: bool = False,
	attention_bias: bool = False,
	attention_dropout: float = 0.0,
	mlp_bias: bool = False,
	head_dim: Optional[int] = None,
	linear_attn_config: Optional[dict] = None,
	**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.pretraining_tp = pretraining_tp
	self.use_cache = use_cache
	self.attention_bias = attention_bias
	self.attention_dropout = attention_dropout
	self.mlp_bias = mlp_bias
	self.head_dim = (
	head_dim
	if head_dim is not None
	else self.hidden_size // self.num_attention_heads
	)

	if linear_attn_config is not None:
	assert linear_attn_config["kda_layers"] is not None
	assert linear_attn_config["full_attn_layers"] is not None
	self.linear_attn_config = linear_attn_config

	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,
	)

	def is_kda_layer(self, layer_idx: int):
	return (
	self.linear_attn_config is not None
	and (layer_idx + 1) in self.linear_attn_config["kda_layers"]
	)


	__all__ = ["ArceeKDAConfig"]