telecomadm1145
/

mamba2_exp3

Text Generation

Model card Files Files and versions

telecomadm1145 commited on Nov 30, 2025

Commit

70e2e50

·

verified ·

1 Parent(s): 7ada0d9

Update config.json

Files changed (1) hide show

config.json +3 -3

config.json CHANGED Viewed

@@ -9,13 +9,13 @@
   "expand": 2,
   "head_dim": 64,
   "hidden_act": "silu",
-  "hidden_size": 512,
   "initializer_range": 0.1,
   "layer_norm_epsilon": 1e-05,
   "model_type": "mamba2",
   "n_groups": 1,
-  "num_heads": 16,
-  "num_hidden_layers": 36,
   "pad_token_id": 0,
   "rescale_prenorm_residual": false,
   "residual_in_fp32": true,

   "expand": 2,
   "head_dim": 64,
   "hidden_act": "silu",
+  "hidden_size": 1024,
   "initializer_range": 0.1,
   "layer_norm_epsilon": 1e-05,
   "model_type": "mamba2",
   "n_groups": 1,
+  "num_heads": 32,
+  "num_hidden_layers": 24,
   "pad_token_id": 0,
   "rescale_prenorm_residual": false,
   "residual_in_fp32": true,