mattdangerw commited on
Commit
7e6fbf6
1 Parent(s): 1fadc70

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. config.json +2 -4
  2. metadata.json +6 -3
  3. tokenizer.json +15 -9
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "module": "keras_nlp.src.models.gpt2.gpt2_backbone",
3
  "class_name": "GPT2Backbone",
4
  "config": {
5
  "name": "gpt2_backbone",
@@ -12,7 +12,5 @@
12
  "dropout": 0.1,
13
  "max_sequence_length": 1024
14
  },
15
- "registered_name": "keras_nlp>GPT2Backbone",
16
- "assets": [],
17
- "weights": "model.weights.h5"
18
  }
 
1
  {
2
+ "module": "keras_hub.src.models.gpt2.gpt2_backbone",
3
  "class_name": "GPT2Backbone",
4
  "config": {
5
  "name": "gpt2_backbone",
 
12
  "dropout": 0.1,
13
  "max_sequence_length": 1024
14
  },
15
+ "registered_name": "keras_hub>GPT2Backbone"
 
 
16
  }
metadata.json CHANGED
@@ -1,6 +1,9 @@
1
  {
2
- "keras_version": "3.0.1",
3
- "keras_nlp_version": "0.7.0",
4
  "parameter_count": 354823168,
5
- "date_saved": "2023-12-27@02:09:15"
 
 
 
6
  }
 
1
  {
2
+ "keras_version": "3.7.0",
3
+ "keras_hub_version": "0.19.0",
4
  "parameter_count": 354823168,
5
+ "date_saved": "2024-12-21@18:42:16",
6
+ "tasks": [
7
+ "CausalLM"
8
+ ]
9
  }
tokenizer.json CHANGED
@@ -1,17 +1,23 @@
1
  {
2
- "module": "keras_nlp.src.models.gpt2.gpt2_tokenizer",
3
  "class_name": "GPT2Tokenizer",
4
  "config": {
5
  "name": "gpt2_tokenizer",
6
  "trainable": true,
7
- "dtype": "int32",
 
 
 
 
 
 
 
 
8
  "sequence_length": null,
9
- "add_prefix_space": false
 
 
 
10
  },
11
- "registered_name": "keras_nlp>GPT2Tokenizer",
12
- "assets": [
13
- "assets/tokenizer/merges.txt",
14
- "assets/tokenizer/vocabulary.json"
15
- ],
16
- "weights": null
17
  }
 
1
  {
2
+ "module": "keras_hub.src.models.gpt2.gpt2_tokenizer",
3
  "class_name": "GPT2Tokenizer",
4
  "config": {
5
  "name": "gpt2_tokenizer",
6
  "trainable": true,
7
+ "dtype": {
8
+ "module": "keras",
9
+ "class_name": "DTypePolicy",
10
+ "config": {
11
+ "name": "int32"
12
+ },
13
+ "registered_name": null
14
+ },
15
+ "config_file": "tokenizer.json",
16
  "sequence_length": null,
17
+ "add_prefix_space": false,
18
+ "unsplittable_tokens": [
19
+ "<|endoftext|>"
20
+ ]
21
  },
22
+ "registered_name": "keras_hub>GPT2Tokenizer"
 
 
 
 
 
23
  }