Divyasreepat
commited on
Commit
•
84af400
1
Parent(s):
69a85da
Upload folder using huggingface_hub
Browse files- README.md +20 -0
- assets/tokenizer/vocabulary.spm +0 -0
- config.json +22 -0
- metadata.json +6 -0
- model.weights.h5 +3 -0
- tokenizer.json +16 -0
README.md
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: keras-hub
|
3 |
+
---
|
4 |
+
This is a [`Mistral` model](https://keras.io/api/keras_hub/models/mistral) uploaded using the KerasHub library and can be used with JAX, TensorFlow, and PyTorch backends.
|
5 |
+
Model config:
|
6 |
+
* **name:** mistral_backbone_1
|
7 |
+
* **trainable:** True
|
8 |
+
* **vocabulary_size:** 32000
|
9 |
+
* **num_layers:** 32
|
10 |
+
* **num_query_heads:** 32
|
11 |
+
* **hidden_dim:** 4096
|
12 |
+
* **intermediate_dim:** 14336
|
13 |
+
* **rope_max_wavelength:** 1000000.0
|
14 |
+
* **rope_scaling_factor:** 1.0
|
15 |
+
* **num_key_value_heads:** 8
|
16 |
+
* **sliding_window:** None
|
17 |
+
* **layer_norm_epsilon:** 1e-05
|
18 |
+
* **dropout:** 0
|
19 |
+
|
20 |
+
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.spm
ADDED
Binary file (493 kB). View file
|
|
config.json
ADDED
@@ -0,0 +1,22 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"module": "keras_nlp.models.mistral.mistral_backbone",
|
3 |
+
"class_name": "MistralBackbone",
|
4 |
+
"config": {
|
5 |
+
"name": "mistral_backbone_1",
|
6 |
+
"trainable": true,
|
7 |
+
"vocabulary_size": 32000,
|
8 |
+
"num_layers": 32,
|
9 |
+
"num_query_heads": 32,
|
10 |
+
"hidden_dim": 4096,
|
11 |
+
"intermediate_dim": 14336,
|
12 |
+
"rope_max_wavelength": 1000000.0,
|
13 |
+
"rope_scaling_factor": 1.0,
|
14 |
+
"num_key_value_heads": 8,
|
15 |
+
"sliding_window": null,
|
16 |
+
"layer_norm_epsilon": 1e-05,
|
17 |
+
"dropout": 0
|
18 |
+
},
|
19 |
+
"registered_name": "keras_nlp>MistralBackbone",
|
20 |
+
"assets": [],
|
21 |
+
"weights": "model.weights.h5"
|
22 |
+
}
|
metadata.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"keras_version": "3.0.5",
|
3 |
+
"keras_nlp_version": "0.9.0",
|
4 |
+
"parameter_count": 7241732096,
|
5 |
+
"date_saved": "2024-03-25@19:58:40"
|
6 |
+
}
|
model.weights.h5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:83e29a61bce2736790afa6479d25b1d6c6a0a5c89d9892cdca786ed5bd49f820
|
3 |
+
size 14484497176
|
tokenizer.json
ADDED
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"module": "keras_nlp.models.mistral.mistral_tokenizer",
|
3 |
+
"class_name": "MistralTokenizer",
|
4 |
+
"config": {
|
5 |
+
"name": "mistral_tokenizer",
|
6 |
+
"trainable": true,
|
7 |
+
"dtype": "int32",
|
8 |
+
"proto": null,
|
9 |
+
"sequence_length": null
|
10 |
+
},
|
11 |
+
"registered_name": "keras_nlp>MistralTokenizer",
|
12 |
+
"assets": [
|
13 |
+
"assets/tokenizer/vocabulary.spm"
|
14 |
+
],
|
15 |
+
"weights": null
|
16 |
+
}
|