breadlicker45 commited on
Commit
612863d
·
1 Parent(s): 8045bc2

Upload config.json

Browse files
Files changed (1) hide show
  1. config.json +10 -16
config.json CHANGED
@@ -1,23 +1,23 @@
1
  {
2
- "_name_or_path": "trained_model/",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
6
  ],
7
  "attn_pdrop": 0.1,
8
- "bos_token_id": 50256,
9
  "embd_pdrop": 0.1,
10
- "eos_token_id": 50256,
11
  "initializer_range": 0.02,
12
  "layer_norm_epsilon": 1e-05,
13
  "line_by_line": true,
14
  "model_type": "gpt2",
15
- "n_ctx": 1024,
16
- "n_embd": 768,
17
- "n_head": 12,
18
  "n_inner": null,
19
- "n_layer": 12,
20
- "n_positions": 1024,
21
  "reorder_and_upcast_attn": false,
22
  "resid_pdrop": 0.1,
23
  "scale_attn_by_inverse_layer_idx": false,
@@ -27,14 +27,8 @@
27
  "summary_proj_to_labels": true,
28
  "summary_type": "cls_index",
29
  "summary_use_proj": true,
30
- "task_specific_params": {
31
- "text-generation": {
32
- "do_sample": true,
33
- "max_length": 50
34
- }
35
- },
36
  "torch_dtype": "float32",
37
- "transformers_version": "4.20.1",
38
  "use_cache": true,
39
- "vocab_size": 50257
40
  }
 
1
  {
2
+ "_name_or_path": "../input/output1/trained_model",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
6
  ],
7
  "attn_pdrop": 0.1,
8
+ "bos_token_id": 0,
9
  "embd_pdrop": 0.1,
10
+ "eos_token_id": 0,
11
  "initializer_range": 0.02,
12
  "layer_norm_epsilon": 1e-05,
13
  "line_by_line": true,
14
  "model_type": "gpt2",
15
+ "n_ctx": 64,
16
+ "n_embd": 128,
17
+ "n_head": 4,
18
  "n_inner": null,
19
+ "n_layer": 4,
20
+ "n_positions": 64,
21
  "reorder_and_upcast_attn": false,
22
  "resid_pdrop": 0.1,
23
  "scale_attn_by_inverse_layer_idx": false,
 
27
  "summary_proj_to_labels": true,
28
  "summary_type": "cls_index",
29
  "summary_use_proj": true,
 
 
 
 
 
 
30
  "torch_dtype": "float32",
31
+ "transformers_version": "4.18.0",
32
  "use_cache": true,
33
+ "vocab_size": 1000
34
  }