RichardErkhov commited on
Commit
3d4f0b9
·
verified ·
1 Parent(s): 4de4dc8

uploaded model

Browse files
Files changed (1) hide show
  1. config.json +94 -0
config.json ADDED
@@ -0,0 +1,94 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Chroma-Powered-RAG-System",
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
+ "do_sample": null,
46
+ "embed_dropout": 0,
47
+ "eos_token_id": 50256,
48
+ "gradient_checkpointing": false,
49
+ "hidden_size": 2048,
50
+ "initializer_range": 0.02,
51
+ "intermediate_size": null,
52
+ "layer_norm_epsilon": 1e-05,
53
+ "max_length": null,
54
+ "max_position_embeddings": 2048,
55
+ "model_type": "gpt_neo",
56
+ "num_heads": 16,
57
+ "num_layers": 24,
58
+ "pad_token_id": 50256,
59
+ "quantization_config": {
60
+ "_load_in_4bit": true,
61
+ "_load_in_8bit": false,
62
+ "bnb_4bit_compute_dtype": "float32",
63
+ "bnb_4bit_quant_storage": "uint8",
64
+ "bnb_4bit_quant_type": "fp4",
65
+ "bnb_4bit_use_double_quant": false,
66
+ "llm_int8_enable_fp32_cpu_offload": false,
67
+ "llm_int8_has_fp16_weight": false,
68
+ "llm_int8_skip_modules": null,
69
+ "llm_int8_threshold": 6.0,
70
+ "load_in_4bit": true,
71
+ "load_in_8bit": false,
72
+ "quant_method": "bitsandbytes"
73
+ },
74
+ "resid_dropout": 0,
75
+ "summary_activation": null,
76
+ "summary_first_dropout": 0.1,
77
+ "summary_proj_to_labels": true,
78
+ "summary_type": "cls_index",
79
+ "summary_use_proj": true,
80
+ "task_specific_params": {
81
+ "text-generation": {
82
+ "do_sample": true,
83
+ "max_length": 50,
84
+ "temperature": 0.9
85
+ }
86
+ },
87
+ "temperature": null,
88
+ "tokenizer_class": "GPT2Tokenizer",
89
+ "torch_dtype": "float16",
90
+ "transformers_version": "4.47.1",
91
+ "use_cache": true,
92
+ "vocab_size": 50257,
93
+ "window_size": 256
94
+ }