Eurdem
/

Bombus_3x8B_v2

Text Generation

Mixture of Experts

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

Eurdem commited on Apr 23

Commit

986eaa7

•

1 Parent(s): 8ab67ef

Update config.json

Files changed (1) hide show

config.json +2 -2

config.json CHANGED Viewed

@@ -1,7 +1,7 @@
 {
   "_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
   "architectures": [
-    "LlamaForCausalLM"
   ],
   "attention_bias": false,
   "attention_dropout": 0.0,
@@ -12,7 +12,7 @@
   "initializer_range": 0.02,
   "intermediate_size": 14336,
   "max_position_embeddings": 8192,
-  "model_type": "llama",
   "num_attention_heads": 32,
   "num_experts_per_tok": 2,
   "num_hidden_layers": 32,

 {
   "_name_or_path": "meta-llama/Meta-Llama-3-8B-Instruct",
   "architectures": [
+    "MixtralForCausalLM"
   ],
   "attention_bias": false,
   "attention_dropout": 0.0,
   "initializer_range": 0.02,
   "intermediate_size": 14336,
   "max_position_embeddings": 8192,
+  "model_type": "mixtral",
   "num_attention_heads": 32,
   "num_experts_per_tok": 2,
   "num_hidden_layers": 32,