sonoisa commited on
Commit
f7c290d
1 Parent(s): 4863222

Add Japanese Sentence-T5 model

Browse files
config.json CHANGED
@@ -1,34 +1,29 @@
1
  {
2
- "_name_or_path": "/content/model/0_T5",
3
  "architectures": [
4
  "T5Model"
5
  ],
6
- "bos_token_id": 0,
7
- "d_ff": 3072,
8
  "d_kv": 64,
9
  "d_model": 768,
10
  "decoder_start_token_id": 0,
11
- "dense_act_fn": "relu",
12
  "dropout_rate": 0.1,
13
  "eos_token_id": 1,
14
- "eos_token_ids": [
15
- 1
16
- ],
17
- "feed_forward_proj": "relu",
18
  "initializer_factor": 1.0,
19
  "is_encoder_decoder": true,
20
- "is_gated_act": false,
21
  "layer_norm_epsilon": 1e-06,
22
- "max_length": 512,
23
  "model_type": "t5",
24
- "n_positions": 512,
25
- "num_beams": 4,
26
  "num_decoder_layers": 12,
27
  "num_heads": 12,
28
  "num_layers": 12,
 
29
  "pad_token_id": 0,
30
  "relative_attention_max_distance": 128,
31
  "relative_attention_num_buckets": 32,
 
32
  "torch_dtype": "float32",
33
  "transformers_version": "4.20.1",
34
  "use_cache": true,
 
1
  {
2
+ "_name_or_path": "/content/drive/MyDrive/T5_models/t5-en+ja_wikipedia-ja_wikipedia-en_cc100-ja_oscar-ja/huggingface_model_800000",
3
  "architectures": [
4
  "T5Model"
5
  ],
6
+ "d_ff": 2048,
 
7
  "d_kv": 64,
8
  "d_model": 768,
9
  "decoder_start_token_id": 0,
10
+ "dense_act_fn": "gelu_new",
11
  "dropout_rate": 0.1,
12
  "eos_token_id": 1,
13
+ "feed_forward_proj": "gated-gelu",
 
 
 
14
  "initializer_factor": 1.0,
15
  "is_encoder_decoder": true,
16
+ "is_gated_act": true,
17
  "layer_norm_epsilon": 1e-06,
 
18
  "model_type": "t5",
 
 
19
  "num_decoder_layers": 12,
20
  "num_heads": 12,
21
  "num_layers": 12,
22
+ "output_past": true,
23
  "pad_token_id": 0,
24
  "relative_attention_max_distance": 128,
25
  "relative_attention_num_buckets": 32,
26
+ "tie_word_embeddings": false,
27
  "torch_dtype": "float32",
28
  "transformers_version": "4.20.1",
29
  "use_cache": true,
config_sentence_transformers.json CHANGED
@@ -2,6 +2,6 @@
2
  "__version__": {
3
  "sentence_transformers": "2.2.2",
4
  "transformers": "4.20.1",
5
- "pytorch": "1.12.0+cu113"
6
  }
7
  }
 
2
  "__version__": {
3
  "sentence_transformers": "2.2.2",
4
  "transformers": "4.20.1",
5
+ "pytorch": "1.11.0+cu113"
6
  }
7
  }
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b9db6254fc4778f70aaa9a37ef488ea03e6ca0b0a01e421da5c699c88a25de3b
3
- size 891700671
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4be43401e983e8c798125b5ccee4f68d2271bee9cfa8d5a3874fadb44161fcf1
3
+ size 891709071
spiece.model CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:da6d60f44af70e494b3762f432efc27b73c0c5e14612153213e6bc496ef60696
3
- size 773857
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df02381552ce4f597ce7bf2685685eacf9222c0d800b2b27ec091886f7e6870f
3
+ size 773638
tokenizer_config.json CHANGED
@@ -103,7 +103,7 @@
103
  ],
104
  "eos_token": "</s>",
105
  "extra_ids": 100,
106
- "name_or_path": "/content/model/0_T5",
107
  "pad_token": "<pad>",
108
  "sp_model_kwargs": {},
109
  "special_tokens_map_file": null,
 
103
  ],
104
  "eos_token": "</s>",
105
  "extra_ids": 100,
106
+ "name_or_path": "/content/drive/MyDrive/T5_models/t5-en+ja_wikipedia-ja_wikipedia-en_cc100-ja_oscar-ja/huggingface_model_800000",
107
  "pad_token": "<pad>",
108
  "sp_model_kwargs": {},
109
  "special_tokens_map_file": null,