Upload MT5ForConditionalGeneration
Browse files- config.json +3 -3
- pytorch_model.bin +2 -2
config.json
CHANGED
|
@@ -7,10 +7,10 @@
|
|
| 7 |
"d_kv": 64,
|
| 8 |
"d_model": 512,
|
| 9 |
"decoder_start_token_id": 0,
|
| 10 |
-
"dense_act_fn": "
|
| 11 |
"dropout_rate": 0.1,
|
| 12 |
"eos_token_id": 1,
|
| 13 |
-
"feed_forward_proj": "gated-
|
| 14 |
"initializer_factor": 1.0,
|
| 15 |
"is_encoder_decoder": true,
|
| 16 |
"is_gated_act": true,
|
|
@@ -23,7 +23,7 @@
|
|
| 23 |
"relative_attention_max_distance": 128,
|
| 24 |
"relative_attention_num_buckets": 32,
|
| 25 |
"scalable_attention": true,
|
| 26 |
-
"tie_word_embeddings":
|
| 27 |
"tokenizer_class": "T5Tokenizer",
|
| 28 |
"torch_dtype": "float32",
|
| 29 |
"transformers_version": "4.31.0.dev0",
|
|
|
|
| 7 |
"d_kv": 64,
|
| 8 |
"d_model": 512,
|
| 9 |
"decoder_start_token_id": 0,
|
| 10 |
+
"dense_act_fn": "relu",
|
| 11 |
"dropout_rate": 0.1,
|
| 12 |
"eos_token_id": 1,
|
| 13 |
+
"feed_forward_proj": "gated-relu",
|
| 14 |
"initializer_factor": 1.0,
|
| 15 |
"is_encoder_decoder": true,
|
| 16 |
"is_gated_act": true,
|
|
|
|
| 23 |
"relative_attention_max_distance": 128,
|
| 24 |
"relative_attention_num_buckets": 32,
|
| 25 |
"scalable_attention": true,
|
| 26 |
+
"tie_word_embeddings": false,
|
| 27 |
"tokenizer_class": "T5Tokenizer",
|
| 28 |
"torch_dtype": "float32",
|
| 29 |
"transformers_version": "4.31.0.dev0",
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:94ee3bfb9c3b281fadb2dbd3e9d4230c9c6d56cb501d25089435eb5e46552aa0
|
| 3 |
+
size 1226475581
|