Divyasreepat
commited on
Commit
•
c0739b9
1
Parent(s):
31d6235
Upload folder using huggingface_hub
Browse files- README.md +19 -0
- assets/tokenizer/merges.txt +0 -0
- assets/tokenizer/vocabulary.json +0 -0
- config.json +19 -0
- metadata.json +6 -0
- model.weights.h5 +3 -0
- tokenizer.json +12 -0
README.md
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: keras-hub
|
3 |
+
---
|
4 |
+
This is a [`Llama3` model](https://keras.io/api/keras_hub/models/llama3) uploaded using the KerasHub library and can be used with JAX, TensorFlow, and PyTorch backends.
|
5 |
+
Model config:
|
6 |
+
* **name:** llama_backbone
|
7 |
+
* **trainable:** True
|
8 |
+
* **vocabulary_size:** 128256
|
9 |
+
* **num_layers:** 32
|
10 |
+
* **num_query_heads:** 32
|
11 |
+
* **hidden_dim:** 4096
|
12 |
+
* **intermediate_dim:** 14336
|
13 |
+
* **rope_max_wavelength:** 500000.0
|
14 |
+
* **rope_scaling_factor:** 1.0
|
15 |
+
* **num_key_value_heads:** 8
|
16 |
+
* **layer_norm_epsilon:** 1e-05
|
17 |
+
* **dropout:** 0
|
18 |
+
|
19 |
+
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/merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
assets/tokenizer/vocabulary.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
config.json
ADDED
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"module": "keras_nlp.src.models.llama3.llama3_backbone",
|
3 |
+
"class_name": "Llama3Backbone",
|
4 |
+
"config": {
|
5 |
+
"name": "llama_backbone",
|
6 |
+
"trainable": true,
|
7 |
+
"vocabulary_size": 128256,
|
8 |
+
"num_layers": 32,
|
9 |
+
"num_query_heads": 32,
|
10 |
+
"hidden_dim": 4096,
|
11 |
+
"intermediate_dim": 14336,
|
12 |
+
"rope_max_wavelength": 500000.0,
|
13 |
+
"rope_scaling_factor": 1.0,
|
14 |
+
"num_key_value_heads": 8,
|
15 |
+
"layer_norm_epsilon": 1e-05,
|
16 |
+
"dropout": 0
|
17 |
+
},
|
18 |
+
"registered_name": "keras_nlp>Llama3Backbone"
|
19 |
+
}
|
metadata.json
ADDED
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"keras_version": "3.0.5",
|
3 |
+
"keras_nlp_version": "0.10.0",
|
4 |
+
"parameter_count": 8030261248,
|
5 |
+
"date_saved": "2024-04-19@23:43:31"
|
6 |
+
}
|
model.weights.h5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:3851648903b142bc41be7c1007cca27fab314f4be2a9d9c7ea03b5c7810efc0f
|
3 |
+
size 16061563672
|
tokenizer.json
ADDED
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"module": "keras_nlp.models.llama3.llama3_tokenizer",
|
3 |
+
"class_name": "Llama3Tokenizer",
|
4 |
+
"config": {
|
5 |
+
"name": "llama3_tokenizer",
|
6 |
+
"trainable": true,
|
7 |
+
"dtype": "int32",
|
8 |
+
"sequence_length": null,
|
9 |
+
"add_prefix_space": false
|
10 |
+
},
|
11 |
+
"registered_name": "keras_nlp>Llama3Tokenizer"
|
12 |
+
}
|