cognitivecomputations
/

Qwen3-58B-Embiggened

Model card Files Files and versions

ehartford commited on 3 days ago

Commit

d191028

·

verified ·

1 Parent(s): beb3696

Update stage1_v2.py

Files changed (1) hide show

stage1_v2.py +1 -1

stage1_v2.py CHANGED Viewed

@@ -46,7 +46,7 @@ HEAD_DIM = 128
 DELTA_HIDDEN = TGT_HIDDEN_SIZE - SRC_HIDDEN_SIZE
 DELTA_INTERMEDIATE = TGT_INTERMEDIATE_SIZE - SRC_INTERMEDIATE_SIZE
-OUTPUT_DIR = "./Qwen3-32B-to-72B-Stage1-v2-sharted"
 # GPU configuration
 NUM_GPUS = 8

 DELTA_HIDDEN = TGT_HIDDEN_SIZE - SRC_HIDDEN_SIZE
 DELTA_INTERMEDIATE = TGT_INTERMEDIATE_SIZE - SRC_INTERMEDIATE_SIZE
+OUTPUT_DIR = "./Qwen3-58B-Embiggened"
 # GPU configuration
 NUM_GPUS = 8