model upload
Browse files- config.json +38 -0
- pytorch_model.bin +3 -0
- tf_model.h5 +3 -0
- tokenizer_config.json +5 -0
- vocab.txt +0 -0
config.json
ADDED
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"activation_dropout": 0.0,
|
3 |
+
"architectures": [
|
4 |
+
"FunnelModel"
|
5 |
+
],
|
6 |
+
"attention_dropout": 0.1,
|
7 |
+
"attention_type": "relative_shift",
|
8 |
+
"block_repeats": [
|
9 |
+
1,
|
10 |
+
1,
|
11 |
+
1
|
12 |
+
],
|
13 |
+
"block_sizes": [
|
14 |
+
6,
|
15 |
+
6,
|
16 |
+
6
|
17 |
+
],
|
18 |
+
"d_head": 64,
|
19 |
+
"d_inner": 3072,
|
20 |
+
"d_model": 768,
|
21 |
+
"embedding_size": 768,
|
22 |
+
"hidden_act": "gelu_new",
|
23 |
+
"hidden_dropout": 0.1,
|
24 |
+
"initializer_range": 0.1,
|
25 |
+
"initializer_std": 0.02,
|
26 |
+
"layer_norm_eps": 1e-09,
|
27 |
+
"max_position_embeddings": 512,
|
28 |
+
"model_type": "funnel",
|
29 |
+
"n_head": 12,
|
30 |
+
"num_decoder_layers": 2,
|
31 |
+
"pool_q_only": true,
|
32 |
+
"pooling_type": "mean",
|
33 |
+
"rel_attn_type": "factorized",
|
34 |
+
"separate_cls": true,
|
35 |
+
"truncate_seq": true,
|
36 |
+
"type_vocab_size": 3,
|
37 |
+
"vocab_size": 42000
|
38 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4aeed1fd83f436ff930b34ff04106dceafd1227a178e01741c737eb04fd97b79
|
3 |
+
size 743572761
|
tf_model.h5
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1f4d470d60aeae04861673166eed1be77bf3548daa4eee3dd4f9112d48f764d5
|
3 |
+
size 743885048
|
tokenizer_config.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"do_lower_case": true,
|
3 |
+
"strip_accents": false,
|
4 |
+
"model_max_length": 512
|
5 |
+
}
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|