Increase context window using rope base freq
Browse files- config.json +2 -2
config.json
CHANGED
|
@@ -16,7 +16,7 @@
|
|
| 16 |
"hidden_size": 3584,
|
| 17 |
"initializer_range": 0.02,
|
| 18 |
"intermediate_size": 14336,
|
| 19 |
-
"max_position_embeddings":
|
| 20 |
"model_type": "gemma2",
|
| 21 |
"num_attention_heads": 16,
|
| 22 |
"num_hidden_layers": 42,
|
|
@@ -24,7 +24,7 @@
|
|
| 24 |
"pad_token_id": 0,
|
| 25 |
"query_pre_attn_scalar": 224,
|
| 26 |
"rms_norm_eps": 1e-06,
|
| 27 |
-
"rope_theta":
|
| 28 |
"sliding_window": 4096,
|
| 29 |
"sliding_window_size": 4096,
|
| 30 |
"torch_dtype": "bfloat16",
|
|
|
|
| 16 |
"hidden_size": 3584,
|
| 17 |
"initializer_range": 0.02,
|
| 18 |
"intermediate_size": 14336,
|
| 19 |
+
"max_position_embeddings": 32768,
|
| 20 |
"model_type": "gemma2",
|
| 21 |
"num_attention_heads": 16,
|
| 22 |
"num_hidden_layers": 42,
|
|
|
|
| 24 |
"pad_token_id": 0,
|
| 25 |
"query_pre_attn_scalar": 224,
|
| 26 |
"rms_norm_eps": 1e-06,
|
| 27 |
+
"rope_theta": 160000.0,
|
| 28 |
"sliding_window": 4096,
|
| 29 |
"sliding_window_size": 4096,
|
| 30 |
"torch_dtype": "bfloat16",
|