Commit
•
9ec770e
1
Parent(s):
a78ac9f
model tuning
Browse files- config.json +6 -5
- pytorch_model.bin +2 -2
config.json
CHANGED
|
@@ -1,19 +1,19 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "t5-base",
|
| 3 |
"architectures": [
|
| 4 |
"T5ForConditionalGeneration"
|
| 5 |
],
|
| 6 |
-
"d_ff":
|
| 7 |
"d_kv": 64,
|
| 8 |
"d_model": 768,
|
| 9 |
"decoder_start_token_id": 0,
|
| 10 |
-
"dense_act_fn": "
|
| 11 |
"dropout_rate": 0.1,
|
| 12 |
"eos_token_id": 1,
|
| 13 |
-
"feed_forward_proj": "
|
| 14 |
"initializer_factor": 1.0,
|
| 15 |
"is_encoder_decoder": true,
|
| 16 |
-
"is_gated_act":
|
| 17 |
"layer_norm_epsilon": 1e-06,
|
| 18 |
"model_type": "t5",
|
| 19 |
"n_positions": 512,
|
|
@@ -53,6 +53,7 @@
|
|
| 53 |
"prefix": "translate English to Romanian: "
|
| 54 |
}
|
| 55 |
},
|
|
|
|
| 56 |
"torch_dtype": "float32",
|
| 57 |
"transformers_version": "4.30.1",
|
| 58 |
"use_cache": true,
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "google/flan-t5-base",
|
| 3 |
"architectures": [
|
| 4 |
"T5ForConditionalGeneration"
|
| 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 |
"n_positions": 512,
|
|
|
|
| 53 |
"prefix": "translate English to Romanian: "
|
| 54 |
}
|
| 55 |
},
|
| 56 |
+
"tie_word_embeddings": false,
|
| 57 |
"torch_dtype": "float32",
|
| 58 |
"transformers_version": "4.30.1",
|
| 59 |
"use_cache": true,
|
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 sha256:224de91c0e9acc62ed17a0ea99208b29fe59a5f222989e10037f518a28456675
|
| 3 |
+
size 990408885
|