Create configuration_bert.py
Browse files- configuration_bert.py +24 -0
configuration_bert.py
ADDED
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
# Copyright 2022 MosaicML Examples authors
|
2 |
+
# SPDX-License-Identifier: Apache-2.0
|
3 |
+
|
4 |
+
from transformers import BertConfig as TransformersBertConfig
|
5 |
+
|
6 |
+
|
7 |
+
class BertConfig(TransformersBertConfig):
|
8 |
+
|
9 |
+
def __init__(
|
10 |
+
self,
|
11 |
+
alibi_starting_size: int = 512,
|
12 |
+
attention_probs_dropout_prob: float = 0.0,
|
13 |
+
**kwargs,
|
14 |
+
):
|
15 |
+
"""Configuration class for MosaicBert.
|
16 |
+
Args:
|
17 |
+
alibi_starting_size (int): Use `alibi_starting_size` to determine how large of an alibi tensor to
|
18 |
+
create when initializing the model. You should be able to ignore this parameter in most cases.
|
19 |
+
Defaults to 512.
|
20 |
+
attention_probs_dropout_prob (float): By default, turn off attention dropout in Mosaic BERT
|
21 |
+
(otherwise, Flash Attention will be off by default). Defaults to 0.0.
|
22 |
+
"""
|
23 |
+
super().__init__(attention_probs_dropout_prob=attention_probs_dropout_prob, **kwargs)
|
24 |
+
self.alibi_starting_size = alibi_starting_size
|