Divyasreepat
commited on
Commit
•
dbea7f4
1
Parent(s):
641c8f5
Upload folder using huggingface_hub
Browse files- README.md +16 -0
- assets/tokenizer/vocabulary.txt +0 -0
- config.json +18 -0
- metadata.json +6 -0
- model.weights.h5 +3 -0
- tokenizer.json +21 -0
README.md
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: keras-hub
|
3 |
+
---
|
4 |
+
This is a [`DistilBert` model](https://keras.io/api/keras_hub/models/distil_bert) uploaded using the KerasHub library and can be used with JAX, TensorFlow, and PyTorch backends.
|
5 |
+
Model config:
|
6 |
+
* **name:** distil_bert_backbone
|
7 |
+
* **trainable:** True
|
8 |
+
* **vocabulary_size:** 30522
|
9 |
+
* **num_layers:** 6
|
10 |
+
* **num_heads:** 12
|
11 |
+
* **hidden_dim:** 768
|
12 |
+
* **intermediate_dim:** 3072
|
13 |
+
* **dropout:** 0.1
|
14 |
+
* **max_sequence_length:** 512
|
15 |
+
|
16 |
+
This model card has been generated automatically and should be completed by the model author. See [Model Cards documentation](https://huggingface.co/docs/hub/model-cards) for more information.
|
assets/tokenizer/vocabulary.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
config.json
ADDED
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"module": "keras_nlp.src.models.distil_bert.distil_bert_backbone",
|
3 |
+
"class_name": "DistilBertBackbone",
|
4 |
+
"config": {
|
5 |
+
"name": "distil_bert_backbone",
|
6 |
+
"trainable": true,
|
7 |
+
"vocabulary_size": 30522,
|
8 |
+
"num_layers": 6,
|
9 |
+
"num_heads": 12,
|
10 |
+
"hidden_dim": 768,
|
11 |
+
"intermediate_dim": 3072,
|
12 |
+
"dropout": 0.1,
|
13 |
+
"max_sequence_length": 512
|
14 |
+
},
|
15 |
+
"registered_name": "keras_nlp>DistilBertBackbone",
|
16 |
+
"assets": [],
|
17 |
+
"weights": "model.weights.h5"
|
18 |
+
}
|
metadata.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"keras_version": "3.0.1",
|
3 |
+
"keras_nlp_version": "0.7.0",
|
4 |
+
"parameter_count": 66362880,
|
5 |
+
"date_saved": "2023-12-27@02:07:24"
|
6 |
+
}
|
model.weights.h5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b084d3272ac49dbf4aebfeeee40efcb2c7b91d95bbfaea896f88d83d5e72dcb9
|
3 |
+
size 265660688
|
tokenizer.json
ADDED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"module": "keras_nlp.src.models.distil_bert.distil_bert_tokenizer",
|
3 |
+
"class_name": "DistilBertTokenizer",
|
4 |
+
"config": {
|
5 |
+
"name": "distil_bert_tokenizer",
|
6 |
+
"trainable": true,
|
7 |
+
"dtype": "int32",
|
8 |
+
"vocabulary": null,
|
9 |
+
"sequence_length": null,
|
10 |
+
"lowercase": true,
|
11 |
+
"strip_accents": false,
|
12 |
+
"split": true,
|
13 |
+
"suffix_indicator": "##",
|
14 |
+
"oov_token": "[UNK]"
|
15 |
+
},
|
16 |
+
"registered_name": "keras_nlp>DistilBertTokenizer",
|
17 |
+
"assets": [
|
18 |
+
"assets/tokenizer/vocabulary.txt"
|
19 |
+
],
|
20 |
+
"weights": null
|
21 |
+
}
|