Add PT checkpoint
Browse files- config.json +3 -1
- pytorch_model.bin +3 -0
config.json
CHANGED
|
@@ -1,4 +1,5 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"LongT5ForConditionalGeneration"
|
| 4 |
],
|
|
@@ -7,6 +8,7 @@
|
|
| 7 |
"d_model": 768,
|
| 8 |
"decoder_start_token_id": 0,
|
| 9 |
"dropout_rate": 0.1,
|
|
|
|
| 10 |
"eos_token_id": 1,
|
| 11 |
"feed_forward_proj": "gated-gelu",
|
| 12 |
"initializer_factor": 1.0,
|
|
@@ -22,8 +24,8 @@
|
|
| 22 |
"pad_token_id": 0,
|
| 23 |
"relative_attention_max_distance": 128,
|
| 24 |
"relative_attention_num_buckets": 32,
|
|
|
|
| 25 |
"transformers_version": "4.19.0.dev0",
|
| 26 |
-
"encoder_attention_type": "local",
|
| 27 |
"use_cache": true,
|
| 28 |
"vocab_size": 32128
|
| 29 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "Stancld/LongT5-Local-Base",
|
| 3 |
"architectures": [
|
| 4 |
"LongT5ForConditionalGeneration"
|
| 5 |
],
|
|
|
|
| 8 |
"d_model": 768,
|
| 9 |
"decoder_start_token_id": 0,
|
| 10 |
"dropout_rate": 0.1,
|
| 11 |
+
"encoder_attention_type": "local",
|
| 12 |
"eos_token_id": 1,
|
| 13 |
"feed_forward_proj": "gated-gelu",
|
| 14 |
"initializer_factor": 1.0,
|
|
|
|
| 24 |
"pad_token_id": 0,
|
| 25 |
"relative_attention_max_distance": 128,
|
| 26 |
"relative_attention_num_buckets": 32,
|
| 27 |
+
"torch_dtype": "float32",
|
| 28 |
"transformers_version": "4.19.0.dev0",
|
|
|
|
| 29 |
"use_cache": true,
|
| 30 |
"vocab_size": 32128
|
| 31 |
}
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f05a759d366d0994af34f8c9be6d1023a917a9290fcc4a42e3d155e0f92ceca8
|
| 3 |
+
size 891705423
|