- config.json +2 -2
config.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
{
|
| 2 |
"architectures": [
|
| 3 |
-
"
|
| 4 |
],
|
| 5 |
"attention_bias": true,
|
| 6 |
"attention_dropout": 0.0,
|
|
@@ -40,7 +40,7 @@
|
|
| 40 |
"full_attention"
|
| 41 |
],
|
| 42 |
"max_position_embeddings": 131072,
|
| 43 |
-
"model_type": "
|
| 44 |
"num_attention_heads": 64,
|
| 45 |
"num_experts_per_tok": 4,
|
| 46 |
"num_hidden_layers": 24,
|
|
|
|
| 1 |
{
|
| 2 |
"architectures": [
|
| 3 |
+
"LexaDeltaForCausalLM"
|
| 4 |
],
|
| 5 |
"attention_bias": true,
|
| 6 |
"attention_dropout": 0.0,
|
|
|
|
| 40 |
"full_attention"
|
| 41 |
],
|
| 42 |
"max_position_embeddings": 131072,
|
| 43 |
+
"model_type": "lexa_delta",
|
| 44 |
"num_attention_heads": 64,
|
| 45 |
"num_experts_per_tok": 4,
|
| 46 |
"num_hidden_layers": 24,
|