sonoisa commited on
Commit
780094b
1 Parent(s): 872f893

Add T5 model and tokenizer pretrained on mC4/ja and Japanese Wikipedia corpus

Browse files
Files changed (3) hide show
  1. config.json +33 -0
  2. pytorch_model.bin +3 -0
  3. spiece.model +3 -0
config.json ADDED
@@ -0,0 +1,33 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "d_ff": 3072,
3
+ "d_kv": 64,
4
+ "d_model": 768,
5
+ "dropout_rate": 0.1,
6
+ "finetuning_task": null,
7
+ "id2label": {
8
+ "0": "LABEL_0",
9
+ "1": "LABEL_1"
10
+ },
11
+ "initializer_factor": 1.0,
12
+ "is_decoder": false,
13
+ "is_encoder_decoder": true,
14
+ "label2id": {
15
+ "LABEL_0": 0,
16
+ "LABEL_1": 1
17
+ },
18
+ "layer_norm_epsilon": 1e-06,
19
+ "n_positions": 512,
20
+ "num_heads": 12,
21
+ "num_labels": 2,
22
+ "num_layers": 12,
23
+ "relative_attention_num_buckets": 32,
24
+ "torchscript": false,
25
+ "use_bfloat16": false,
26
+ "vocab_size": 32128,
27
+ "max_length": 512,
28
+ "num_beams": 4,
29
+ "decoder_start_token_id": 0,
30
+ "pad_token_id": 0,
31
+ "bos_token_id": 0,
32
+ "eos_token_ids": [1]
33
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7961b2157dabf06ec5f3ee8ede9fd331e32b6adeb66245320418e5394b1c5444
3
+ size 891727167
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:575bba4012fcabb11c89be62a45afdcb2ef648e880f021703b96a8a173f6b3b7
3
+ size 804231