ai-forever
commited on
Commit
•
8c2dc46
1
Parent(s):
76a68e8
Upload T5ForConditionalGeneration
Browse files- config.json +33 -0
- generation_config.json +8 -0
- model.safetensors +3 -0
config.json
ADDED
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "../flan_t5_small_2",
|
3 |
+
"architectures": [
|
4 |
+
"T5ForConditionalGeneration"
|
5 |
+
],
|
6 |
+
"bos_token_id": 1,
|
7 |
+
"classifier_dropout": 0.0,
|
8 |
+
"d_ff": 1024,
|
9 |
+
"d_kv": 64,
|
10 |
+
"d_model": 512,
|
11 |
+
"decoder_start_token_id": 0,
|
12 |
+
"dense_act_fn": "gelu_new",
|
13 |
+
"dropout_rate": 0.1,
|
14 |
+
"eos_token_id": 2,
|
15 |
+
"feed_forward_proj": "gated-gelu",
|
16 |
+
"initializer_factor": 1.0,
|
17 |
+
"is_encoder_decoder": true,
|
18 |
+
"is_gated_act": true,
|
19 |
+
"layer_norm_epsilon": 1e-06,
|
20 |
+
"model_type": "t5",
|
21 |
+
"num_decoder_layers": 8,
|
22 |
+
"num_heads": 6,
|
23 |
+
"num_layers": 8,
|
24 |
+
"output_past": true,
|
25 |
+
"pad_token_id": 0,
|
26 |
+
"relative_attention_max_distance": 128,
|
27 |
+
"relative_attention_num_buckets": 32,
|
28 |
+
"tie_word_embeddings": false,
|
29 |
+
"torch_dtype": "float32",
|
30 |
+
"transformers_version": "4.38.2",
|
31 |
+
"use_cache": true,
|
32 |
+
"vocab_size": 50364
|
33 |
+
}
|
generation_config.json
ADDED
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_from_model_config": true,
|
3 |
+
"bos_token_id": 1,
|
4 |
+
"decoder_start_token_id": 0,
|
5 |
+
"eos_token_id": 2,
|
6 |
+
"pad_token_id": 0,
|
7 |
+
"transformers_version": "4.38.2"
|
8 |
+
}
|
model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:33d3defd12154c821525b257febb0241c6ae02e5e834802c18f9e787e00ff6fa
|
3 |
+
size 382561704
|