dropout 0.3
Browse files- config.json +4 -4
config.json
CHANGED
|
@@ -1,15 +1,15 @@
|
|
| 1 |
{
|
| 2 |
"_name_or_path": "bart-base",
|
| 3 |
-
"activation_dropout": 0.
|
| 4 |
"activation_function": "gelu",
|
| 5 |
"add_bias_logits": false,
|
| 6 |
"add_final_layer_norm": false,
|
| 7 |
"architectures": [
|
| 8 |
"BartModel"
|
| 9 |
],
|
| 10 |
-
"attention_dropout": 0.
|
| 11 |
"bos_token_id": 0,
|
| 12 |
-
"classif_dropout": 0.
|
| 13 |
"classifier_dropout": 0.0,
|
| 14 |
"d_model": 512,
|
| 15 |
"decoder_attention_heads": 8,
|
|
@@ -17,7 +17,7 @@
|
|
| 17 |
"decoder_layerdrop": 0.0,
|
| 18 |
"decoder_layers": 6,
|
| 19 |
"decoder_start_token_id": 2,
|
| 20 |
-
"dropout": 0.
|
| 21 |
"early_stopping": true,
|
| 22 |
"encoder_attention_heads": 8,
|
| 23 |
"encoder_ffn_dim": 2048,
|
|
|
|
| 1 |
{
|
| 2 |
"_name_or_path": "bart-base",
|
| 3 |
+
"activation_dropout": 0.3,
|
| 4 |
"activation_function": "gelu",
|
| 5 |
"add_bias_logits": false,
|
| 6 |
"add_final_layer_norm": false,
|
| 7 |
"architectures": [
|
| 8 |
"BartModel"
|
| 9 |
],
|
| 10 |
+
"attention_dropout": 0.3,
|
| 11 |
"bos_token_id": 0,
|
| 12 |
+
"classif_dropout": 0.3,
|
| 13 |
"classifier_dropout": 0.0,
|
| 14 |
"d_model": 512,
|
| 15 |
"decoder_attention_heads": 8,
|
|
|
|
| 17 |
"decoder_layerdrop": 0.0,
|
| 18 |
"decoder_layers": 6,
|
| 19 |
"decoder_start_token_id": 2,
|
| 20 |
+
"dropout": 0.3,
|
| 21 |
"early_stopping": true,
|
| 22 |
"encoder_attention_heads": 8,
|
| 23 |
"encoder_ffn_dim": 2048,
|