RichardErkhov commited on
Commit
62763e7
·
verified ·
1 Parent(s): be7ad0d

uploaded model

Browse files
Files changed (1) hide show
  1. config.json +90 -0
config.json ADDED
@@ -0,0 +1,90 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "GPT-Neo-1.3B-Adventure",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPTNeoForCausalLM"
6
+ ],
7
+ "attention_dropout": 0,
8
+ "attention_layers": [
9
+ "global",
10
+ "local",
11
+ "global",
12
+ "local",
13
+ "global",
14
+ "local",
15
+ "global",
16
+ "local",
17
+ "global",
18
+ "local",
19
+ "global",
20
+ "local",
21
+ "global",
22
+ "local",
23
+ "global",
24
+ "local",
25
+ "global",
26
+ "local",
27
+ "global",
28
+ "local",
29
+ "global",
30
+ "local",
31
+ "global",
32
+ "local"
33
+ ],
34
+ "attention_types": [
35
+ [
36
+ [
37
+ "global",
38
+ "local"
39
+ ],
40
+ 12
41
+ ]
42
+ ],
43
+ "bos_token_id": 50256,
44
+ "classifier_dropout": 0.1,
45
+ "embed_dropout": 0,
46
+ "eos_token_id": 50256,
47
+ "hidden_size": 2048,
48
+ "initializer_range": 0.02,
49
+ "intermediate_size": null,
50
+ "layer_norm_epsilon": 1e-05,
51
+ "max_position_embeddings": 2048,
52
+ "model_type": "gpt_neo",
53
+ "num_heads": 16,
54
+ "num_layers": 24,
55
+ "quantization_config": {
56
+ "_load_in_4bit": false,
57
+ "_load_in_8bit": true,
58
+ "bnb_4bit_compute_dtype": "float32",
59
+ "bnb_4bit_quant_storage": "uint8",
60
+ "bnb_4bit_quant_type": "fp4",
61
+ "bnb_4bit_use_double_quant": false,
62
+ "llm_int8_enable_fp32_cpu_offload": false,
63
+ "llm_int8_has_fp16_weight": false,
64
+ "llm_int8_skip_modules": null,
65
+ "llm_int8_threshold": 6.0,
66
+ "load_in_4bit": false,
67
+ "load_in_8bit": true,
68
+ "quant_method": "bitsandbytes"
69
+ },
70
+ "resid_dropout": 0,
71
+ "summary_activation": null,
72
+ "summary_first_dropout": 0.1,
73
+ "summary_proj_to_labels": true,
74
+ "summary_type": "cls_index",
75
+ "summary_use_proj": true,
76
+ "task_specific_params": {
77
+ "text-generation": {
78
+ "do_sample": true,
79
+ "max_length": 50,
80
+ "temperature": 0.9
81
+ }
82
+ },
83
+ "tokenizer_class": "GPT2Tokenizer",
84
+ "torch_dtype": "float16",
85
+ "transformers_version": "4.47.1",
86
+ "use_cache": false,
87
+ "vocab_size": 50257,
88
+ "welcome": "You are currently running adventure model `CYS 1.3B`\n\n This model is made by [Mr. Seeker](https://www.patreon.com/mrseeker)\n\n### How to use this model\n\nAdventure models are designed to generate CYOA (choose your own adventure) stories, similar to Kings Quest. Use the authors note to give it a certain genre to follow, use memory to give an overview of the story and use World Information to give it specific details about the characters. Make sure you are running in adventure mode for full effect. To start off, give the AI a prompt in the \"You\" form, for example: You look around the building.",
89
+ "window_size": 256
90
+ }