Text Generation
Transformers
PyTorch
gptj
Inference Endpoints
juliensalinas commited on
Commit
f57445f
1 Parent(s): cde480b

Upload config.json with huggingface_hub

Browse files
Files changed (1) hide show
  1. config.json +13 -26
config.json CHANGED
@@ -1,43 +1,27 @@
1
  {
2
  "activation_function": "gelu_new",
3
  "architectures": [
4
- "GPTNeoForCausalLM"
5
- ],
6
- "attention_dropout": 0,
7
- "attention_layers": ["global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global"],
8
- "attention_types": [
9
- [
10
- [
11
- "global"
12
- ],
13
- 28
14
- ]
15
  ],
 
16
  "bos_token_id": 50256,
17
- "embed_dropout": 0,
18
  "eos_token_id": 50256,
19
  "gradient_checkpointing": false,
20
- "hidden_size": 4096,
21
  "initializer_range": 0.02,
22
- "intermediate_size": null,
23
- "jax": true,
24
  "layer_norm_epsilon": 1e-05,
25
- "max_position_embeddings": 2048,
26
- "model_type": "gpt_neo",
27
- "num_heads": 16,
28
- "num_layers": 28,
29
- "resid_dropout": 0,
30
- "rotary": true,
31
  "rotary_dim": 64,
32
  "summary_activation": null,
33
  "summary_first_dropout": 0.1,
34
  "summary_proj_to_labels": true,
35
  "summary_type": "cls_index",
36
  "summary_use_proj": true,
37
- "transformers_version": "4.5.0.dev0",
38
- "use_cache": true,
39
- "vocab_size": 50400,
40
- "window_size": 256,
41
  "tokenizer_class": "GPT2Tokenizer",
42
  "task_specific_params": {
43
  "text-generation": {
@@ -45,5 +29,8 @@
45
  "temperature": 1.0,
46
  "max_length": 50
47
  }
48
- }
 
 
 
49
  }
 
1
  {
2
  "activation_function": "gelu_new",
3
  "architectures": [
4
+ "GPTJForCausalLM"
 
 
 
 
 
 
 
 
 
 
5
  ],
6
+ "attn_pdrop": 0.0,
7
  "bos_token_id": 50256,
8
+ "embd_pdrop": 0.0,
9
  "eos_token_id": 50256,
10
  "gradient_checkpointing": false,
 
11
  "initializer_range": 0.02,
 
 
12
  "layer_norm_epsilon": 1e-05,
13
+ "model_type": "gptj",
14
+ "n_embd": 4096,
15
+ "n_head": 16,
16
+ "n_layer": 28,
17
+ "n_positions": 2048,
 
18
  "rotary_dim": 64,
19
  "summary_activation": null,
20
  "summary_first_dropout": 0.1,
21
  "summary_proj_to_labels": true,
22
  "summary_type": "cls_index",
23
  "summary_use_proj": true,
24
+ "transformers_version": "4.10.0.dev0",
 
 
 
25
  "tokenizer_class": "GPT2Tokenizer",
26
  "task_specific_params": {
27
  "text-generation": {
 
29
  "temperature": 1.0,
30
  "max_length": 50
31
  }
32
+ },
33
+ "torch_dtype": "float16",
34
+ "use_cache": true,
35
+ "vocab_size": 50400
36
  }