# Copyright 2025 The EasyDeL Author @erfanzar (Erfan Zare Chavoshi).
#
# 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
#
# https://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.
import typing
from eformer.common_types import ColumnWise, Replicated, RowWise
from easydel.infra.base_module import EasyDeLBaseConfig
from easydel.infra.etils import EasyDeLGradientCheckPointers
from easydel.infra.factory import register_config
[docs]@register_config("gpt2")
class GPT2Config(EasyDeLBaseConfig):
"""
Configuration objects inherit from [`EasyDeLBaseConfig`] and can be used to control the model outputs. Read
the documentation from [`EasyDeLBaseConfig`] for more information.
Args:
vocab_size (`int`, *optional*, defaults to 50257):
Vocabulary size of the GPT-2 model. Defines the number of different tokens that can be represented by the
`inputs_ids` passed to the forward method.
n_positions (`int`, *optional*, defaults to 1024):
The maximum sequence length that this model might ever be used with. Typically set this to something large
just in case (e.g., 2048 or 4096).
n_embd (`int`, *optional*, defaults to 768):
Dimensionality of the encoder layers and the pooler layer.
n_layer (`int`, *optional*, defaults to 12):
Number of hidden layers in the Transformer encoder.
n_head (`int`, *optional*, defaults to 12):
Number of attention heads for each attention layer in the Transformer encoder.
n_inner (`int`, *optional*):
Dimensionality of the inner feed-forward layers.
activation_function (`str`, *optional*, defaults to `"gelu_new"`):
The non-linear activation function (function or string) to use in the encoder and pooler. If string,
`"gelu"`, `"relu"`, `"swish"` and `"gelu_new"` are supported.
resid_pdrop (`float`, *optional*, defaults to 0.1):
The dropout probability for all fully connected layers in the embeddings, encoder, and pooler.
embd_pdrop (`float`, *optional*, defaults to 0.1):
The dropout ratio for the embeddings.
attn_pdrop (`float`, *optional*, defaults to 0.1):
The dropout ratio for the attention probabilities.
layer_norm_epsilon (`float`, *optional*, defaults to 1e-5):
The epsilon to use in the layer normalization layers.
initializer_range (`float`, *optional*, defaults to 0.02):
The standard deviation of the truncated_normal_initializer for initializing all weight matrices.
summary_type (`str`, *optional*, defaults to `"cls_index"`):
The summary type to use. Possible values are `"cls_index"` (equivalent to the output of the last token
of the first sentence in a sequence) and `"last"` (equivalent to the output of the last token in
the sequence).
summary_use_proj (`bool`, *optional*, defaults to `True`):
Whether to use a projection after the vector extraction.
summary_activation (`str`, *optional*):
The activation to use for the summary.
summary_proj_to_labels (`bool`, *optional*, defaults to `True`):
Whether to project the summary to the labels.
summary_first_dropout (`float`, *optional*, defaults to 0.1):
The dropout ratio to be used after the projection and activation.
scale_attn_weights (`bool`, *optional*, defaults to `True`):
Scale attention weights by dividing by sqrt(hidden_size).
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`.
bos_token_id (`int`, *optional*, defaults to 50256):
The id of the *beginning-of-sequence* token.
eos_token_id (`int`, *optional*, defaults to 50256):
The id of the *end-of-sequence* token.
scale_attn_by_inverse_layer_idx (`bool`, *optional*, defaults to `False`):
Whether to scale attention weights by `1 / layer_idx + 1`.
reorder_and_upcast_attn (`bool`, *optional*, defaults to `False`):
Whether to reorder and upcast attention.
gradient_checkpointing (`str`, *optional*, defaults to `"nothing_saveable"`):
The gradient checkpointing configuration.
tie_word_embeddings (`bool`, *optional*, defaults to `False`):
Whether to tie the weights of the input embeddings and the output embeddings.
bits (`int`, *optional*):
The number of bits to quantize the model to.
"""
model_type: str = "gpt2"
keys_to_ignore_at_inference: typing.ClassVar = ["past_key_values"]
attribute_map: typing.ClassVar = {
"hidden_size": "n_embd",
"max_position_embeddings": "n_positions",
"num_attention_heads": "n_head",
"num_hidden_layers": "n_layer",
}
def __init__(
self,
vocab_size=50257,
n_positions=1024,
n_embd=768,
n_layer=12,
n_head=12,
n_inner=None,
activation_function="gelu_new",
resid_pdrop=0.1,
embd_pdrop=0.1,
attn_pdrop=0.1,
layer_norm_epsilon=1e-5,
initializer_range=0.02,
summary_type="cls_index",
summary_use_proj=True,
summary_activation=None,
summary_proj_to_labels=True,
summary_first_dropout=0.1,
scale_attn_weights=True,
use_cache=True,
bos_token_id=50256,
eos_token_id=50256,
scale_attn_by_inverse_layer_idx=False,
reorder_and_upcast_attn=False,
gradient_checkpointing: EasyDeLGradientCheckPointers = EasyDeLGradientCheckPointers.NONE,
tie_word_embeddings: bool = False,
bits: int | None = None,
**kwargs,
):
"""Initializes a GPT2Config object.
Args:
vocab_size (int, optional): Vocabulary size. Defaults to 50257.
n_positions (int, optional): Maximum sequence length. Defaults to 1024.
n_embd (int, optional): Hidden size. Defaults to 768.
n_layer (int, optional): Number of hidden layers. Defaults to 12.
n_head (int, optional): Number of attention heads. Defaults to 12.
n_inner (int, optional): Inner dimension of FFN. Defaults to None.
activation_function (str, optional): Activation function. Defaults to "gelu_new".
resid_pdrop (float, optional): Residual dropout probability. Defaults to 0.1.
embd_pdrop (float, optional): Embedding dropout probability. Defaults to 0.1.
attn_pdrop (float, optional): Attention dropout probability. Defaults to 0.1.
layer_norm_epsilon (float, optional): Epsilon for layer normalization. Defaults to 1e-5.
initializer_range (float, optional): Initializer range. Defaults to 0.02.
summary_type (str, optional): Type of summary. Defaults to "cls_index".
summary_use_proj (bool, optional): Whether to use projection in summary. Defaults to True.
summary_activation (str, optional): Activation for summary. Defaults to None.
summary_proj_to_labels (bool, optional): Whether to project summary to labels. Defaults to True.
summary_first_dropout (float, optional): Dropout after summary projection. Defaults to 0.1.
scale_attn_weights (bool, optional): Whether to scale attention weights. Defaults to True.
use_cache (bool, optional): Whether to use KV cache. Defaults to True.
bos_token_id (int, optional): Beginning-of-sequence token ID. Defaults to 50256.
eos_token_id (int, optional): End-of-sequence token ID. Defaults to 50256.
scale_attn_by_inverse_layer_idx (bool, optional):
Whether to scale attention by inverse layer index. Defaults to False.
reorder_and_upcast_attn (bool, optional): Whether to reorder and upcast attention. Defaults to False.
gradient_checkpointing (EasyDeLGradientCheckPointers, optional):
Gradient checkpointing strategy. Defaults to EasyDeLGradientCheckPointers.NONE.
tie_word_embeddings (bool, optional): Whether to tie input/output embeddings. Defaults to False.
bits (tp.Optional[int], optional): Quantization bits. Defaults to None.
**kwargs: Additional keyword arguments.
"""
self.vocab_size = vocab_size
self.n_positions = n_positions
self.n_embd = n_embd
self.n_layer = n_layer
self.n_head = n_head
self.n_inner = n_inner
self.activation_function = activation_function
self.resid_pdrop = resid_pdrop
self.embd_pdrop = embd_pdrop
self.attn_pdrop = attn_pdrop
self.layer_norm_epsilon = layer_norm_epsilon
self.initializer_range = initializer_range
self.summary_type = summary_type
self.summary_use_proj = summary_use_proj
self.summary_activation = summary_activation
self.summary_first_dropout = summary_first_dropout
self.summary_proj_to_labels = summary_proj_to_labels
self.scale_attn_weights = scale_attn_weights
self.use_cache = use_cache
self.scale_attn_by_inverse_layer_idx = scale_attn_by_inverse_layer_idx
self.reorder_and_upcast_attn = reorder_and_upcast_attn
self.bos_token_id = bos_token_id
self.eos_token_id = eos_token_id
self.gradient_checkpointing = gradient_checkpointing
self.bits = bits
super().__init__(
bos_token_id=bos_token_id,
eos_token_id=eos_token_id,
tie_word_embeddings=tie_word_embeddings,
bits=bits,
**kwargs,
)
[docs] def get_partition_rules(self, *args, **kwargs):
"""
Get the partition rules for the model.
Returns:
`tp.Tuple[tp.Tuple[str, PartitionSpec]]`: The partition rules.
"""
pmag = self.partition_manager
return (
(r"wte/embedding", pmag.resolve(ColumnWise)),
(r"wpe/embedding", pmag.resolve(Replicated)),
(r"(attn|crossattention)/c_attn/kernel", pmag.resolve(ColumnWise)),
(r"(attn|crossattention)/q_attn/kernel", pmag.resolve(ColumnWise)),
(r"(attn|crossattention)/c_proj/kernel", pmag.resolve(RowWise)),
(r"mlp/c_fc/kernel", pmag.resolve(ColumnWise)),
(r"mlp/c_proj/kernel", pmag.resolve(RowWise)),
(r".*/(ln_1|ln_2|ln_cross_attn|ln_f)/scale", pmag.resolve(Replicated)),
(r".*/(ln_1|ln_2|ln_cross_attn|ln_f)/bias", pmag.resolve(Replicated)),
(r"lm_head/kernel", pmag.resolve(ColumnWise)),
(r".*(c_attn|q_attn|c_proj|c_fc|lm_head)/bias", pmag.resolve(Replicated)),
(r".*", pmag.resolve(Replicated)),
)