aapot
commited on
Commit
•
2572326
1
Parent(s):
9025e0d
Add Pytorch model
Browse files- config.json +5 -1
- pytorch_model.bin +3 -0
config.json
CHANGED
@@ -1,4 +1,5 @@
|
|
1 |
{
|
|
|
2 |
"architectures": [
|
3 |
"T5ForConditionalGeneration"
|
4 |
],
|
@@ -6,12 +7,14 @@
|
|
6 |
"d_kv": 64,
|
7 |
"d_model": 1536,
|
8 |
"decoder_start_token_id": 0,
|
|
|
9 |
"dropout_rate": 0.1,
|
10 |
"eos_token_id": 1,
|
11 |
"feed_forward_proj": "gated-gelu",
|
12 |
"gradient_checkpointing": false,
|
13 |
"initializer_factor": 1.0,
|
14 |
"is_encoder_decoder": true,
|
|
|
15 |
"layer_norm_epsilon": 1e-06,
|
16 |
"model_type": "t5",
|
17 |
"num_decoder_layers": 6,
|
@@ -23,7 +26,8 @@
|
|
23 |
"relative_attention_num_buckets": 32,
|
24 |
"tie_word_embeddings": false,
|
25 |
"tokenizer_class": "ByT5Tokenizer",
|
26 |
-
"
|
|
|
27 |
"use_cache": true,
|
28 |
"vocab_size": 384
|
29 |
}
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "Finnish-NLP/byt5-base-finnish",
|
3 |
"architectures": [
|
4 |
"T5ForConditionalGeneration"
|
5 |
],
|
|
|
7 |
"d_kv": 64,
|
8 |
"d_model": 1536,
|
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 |
"gradient_checkpointing": false,
|
15 |
"initializer_factor": 1.0,
|
16 |
"is_encoder_decoder": true,
|
17 |
+
"is_gated_act": true,
|
18 |
"layer_norm_epsilon": 1e-06,
|
19 |
"model_type": "t5",
|
20 |
"num_decoder_layers": 6,
|
|
|
26 |
"relative_attention_num_buckets": 32,
|
27 |
"tie_word_embeddings": false,
|
28 |
"tokenizer_class": "ByT5Tokenizer",
|
29 |
+
"torch_dtype": "float32",
|
30 |
+
"transformers_version": "4.20.1",
|
31 |
"use_cache": true,
|
32 |
"vocab_size": 384
|
33 |
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5c2205800cf5fefd1601751d9f0a7c65f27619df9e7ab547c4010090d832bac3
|
3 |
+
size 2326694409
|