Commit
·
5d31253
1
Parent(s):
01cc289
Update config.json
Browse filesinsert the config from https://huggingface.co./benjamin/gerpt2/blob/main/config.json
- config.json +7 -12
config.json
CHANGED
@@ -1,13 +1,14 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "stefan-it/german-gpt2-larger",
|
3 |
"activation_function": "gelu_new",
|
4 |
"architectures": [
|
5 |
"GPT2LMHeadModel"
|
6 |
],
|
7 |
-
"
|
|
|
8 |
"bos_token_id": 50256,
|
9 |
-
"embd_pdrop": 0.
|
10 |
"eos_token_id": 50256,
|
|
|
11 |
"gradient_checkpointing": false,
|
12 |
"initializer_range": 0.02,
|
13 |
"layer_norm_epsilon": 1e-05,
|
@@ -18,10 +19,7 @@
|
|
18 |
"n_inner": null,
|
19 |
"n_layer": 12,
|
20 |
"n_positions": 1024,
|
21 |
-
"
|
22 |
-
"resid_pdrop": 0.0,
|
23 |
-
"scale_attn_by_inverse_layer_idx": false,
|
24 |
-
"scale_attn_weights": true,
|
25 |
"summary_activation": null,
|
26 |
"summary_first_dropout": 0.1,
|
27 |
"summary_proj_to_labels": true,
|
@@ -30,11 +28,8 @@
|
|
30 |
"task_specific_params": {
|
31 |
"text-generation": {
|
32 |
"do_sample": true,
|
33 |
-
"max_length":
|
34 |
}
|
35 |
},
|
36 |
-
"
|
37 |
-
"transformers_version": "4.20.1",
|
38 |
-
"use_cache": true,
|
39 |
-
"vocab_size": 50265
|
40 |
}
|
|
|
1 |
{
|
|
|
2 |
"activation_function": "gelu_new",
|
3 |
"architectures": [
|
4 |
"GPT2LMHeadModel"
|
5 |
],
|
6 |
+
"tie_word_embeddings": false,
|
7 |
+
"attn_pdrop": 0.1,
|
8 |
"bos_token_id": 50256,
|
9 |
+
"embd_pdrop": 0.1,
|
10 |
"eos_token_id": 50256,
|
11 |
+
"pad_token_id": 0,
|
12 |
"gradient_checkpointing": false,
|
13 |
"initializer_range": 0.02,
|
14 |
"layer_norm_epsilon": 1e-05,
|
|
|
19 |
"n_inner": null,
|
20 |
"n_layer": 12,
|
21 |
"n_positions": 1024,
|
22 |
+
"resid_pdrop": 0.1,
|
|
|
|
|
|
|
23 |
"summary_activation": null,
|
24 |
"summary_first_dropout": 0.1,
|
25 |
"summary_proj_to_labels": true,
|
|
|
28 |
"task_specific_params": {
|
29 |
"text-generation": {
|
30 |
"do_sample": true,
|
31 |
+
"max_length": 100
|
32 |
}
|
33 |
},
|
34 |
+
"vocab_size": 50257
|
|
|
|
|
|
|
35 |
}
|