set attention_head_dim to 2

config.json

@@ -5,7 +5,7 @@
     "Zamba2ForCausalLM"
   ],
   "attention_dropout": 0.0,
-  "attention_head_dim": 1,
+  "attention_head_dim": 2,
   "attention_hidden_size": 32,
   "bos_token_id": 1,
   "chunk_size": 256,
@@ -54,7 +54,11 @@
   "time_step_limit": null,
   "time_step_max": 0.1,
   "time_step_min": 0.001,
+<<<<<<< Updated upstream
   "transformers_version": "5.0.0",
+=======
+  "transformers_version": "4.57.6",
+>>>>>>> Stashed changes
   "use_cache": true,
   "use_conv_bias": true,
   "use_long_context": false,

generation_config.json

@@ -3,5 +3,9 @@
   "bos_token_id": 1,
   "eos_token_id": 2,
   "pad_token_id": 0,
+<<<<<<< Updated upstream
   "transformers_version": "5.0.0"
+=======
+  "transformers_version": "4.57.6"
+>>>>>>> Stashed changes
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:257244fe95a815ccdfc00b230826eb464c8b6bfc791d6e8c35ec7a0e470ae4fb
-size 3398168
+oid sha256:a59661b1a44705c94679f15edadb04c60a80eb324d27fd0f81db8ee513779c78
+size 3412504