kakaocorp
/

kanana-1.5-15.7b-a3b-base

Text Generation

text-generation-inference

Model card Files Files and versions Community

lana-ny commited on Jun 11

Commit

f47a5d8

·

verified ·

1 Parent(s): d1e75fe

Update config.json

Files changed (1) hide show

config.json +1 -1

config.json CHANGED Viewed

@@ -20,7 +20,7 @@
   "output_router_logits": false,
   "rms_norm_eps": 1e-05,
   "rope_theta": 16000000.0,
-  "router_aux_loss_coef": 0.001,
   "router_jitter_noise": 0.0,
   "scoring_func": "softmax",
   "sliding_window": null,

   "output_router_logits": false,
   "rms_norm_eps": 1e-05,
   "rope_theta": 16000000.0,
+  "router_aux_loss_coef": 0.005,
   "router_jitter_noise": 0.0,
   "scoring_func": "softmax",
   "sliding_window": null,