Update stage1_v2.py
Browse files- stage1_v2.py +1 -1
stage1_v2.py
CHANGED
|
@@ -46,7 +46,7 @@ HEAD_DIM = 128
|
|
| 46 |
DELTA_HIDDEN = TGT_HIDDEN_SIZE - SRC_HIDDEN_SIZE
|
| 47 |
DELTA_INTERMEDIATE = TGT_INTERMEDIATE_SIZE - SRC_INTERMEDIATE_SIZE
|
| 48 |
|
| 49 |
-
OUTPUT_DIR = "./Qwen3-
|
| 50 |
|
| 51 |
# GPU configuration
|
| 52 |
NUM_GPUS = 8
|
|
|
|
| 46 |
DELTA_HIDDEN = TGT_HIDDEN_SIZE - SRC_HIDDEN_SIZE
|
| 47 |
DELTA_INTERMEDIATE = TGT_INTERMEDIATE_SIZE - SRC_INTERMEDIATE_SIZE
|
| 48 |
|
| 49 |
+
OUTPUT_DIR = "./Qwen3-58B-Embiggened"
|
| 50 |
|
| 51 |
# GPU configuration
|
| 52 |
NUM_GPUS = 8
|