UCLA-AGI
/

Gemma-2-9B-It-SPPO-Iter3

Text Generation

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

Fix query_pre_attn_scalar value

#6

by Dampfinchen - opened Jul 14

base: refs/heads/main

←

from: refs/pr/6

Discussion Files changed

Files changed (1) hide show

config.json +1 -1

config.json CHANGED Viewed

@@ -22,7 +22,7 @@
   "num_hidden_layers": 42,
   "num_key_value_heads": 8,
   "pad_token_id": 0,
-  "query_pre_attn_scalar": 224,
   "rms_norm_eps": 1e-06,
   "rope_theta": 10000.0,
   "sliding_window": 4096,

   "num_hidden_layers": 42,
   "num_key_value_heads": 8,
   "pad_token_id": 0,
+  "query_pre_attn_scalar": 256,
   "rms_norm_eps": 1e-06,
   "rope_theta": 10000.0,
   "sliding_window": 4096,