Upload model

config.json

@@ -6,15 +6,29 @@
     "AutoConfig": "configuration_musilingo.MusiLingoConfig",
     "AutoModel": "modelling_musilingo.MusilingoModel"
   },
+  "bos_token_id": 1,
   "device_8bit": 0,
   "end_sym": "\n",
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
   "llama_model": "lmsys/vicuna-7b-delta-v0",
   "low_resource": false,
+  "max_position_embeddings": 2048,
   "max_txt_len": 32,
   "mert_model": "m-a-p/MERT-v1-330M",
   "model_type": "musilingo",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "pad_token_id": 0,
   "prompt_path": "",
   "prompt_template": "###Human: {} ###Assistant: ",
+  "rms_norm_eps": 1e-06,
+  "tie_word_embeddings": false,
   "torch_dtype": "float32",
-  "transformers_version": "4.39.3"
+  "transformers_version": "4.39.3",
+  "use_cache": true,
+  "vocab_size": 32001
 }

model-00001-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c24790aba36855d1eb25095c68b7a3a782d09c5281d1ddebc788f6946c374b3b
+oid sha256:b3dfb974964f6b723d558e3107ee5b98ae1f13a08e681dfe7106806a01078e74
 size 4986465504

model-00002-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:81c614ec56b4947a29b7f48a1e0d557e0a12ef70cf626194da05c02a1cfb85ff
+oid sha256:54cb40c370505a6378fa41c4a30c352df90599ce458f55e36716d874cf292c31
 size 4947397256

model-00003-of-00003.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:61092e0edb4cb84d0a39493c41394755bd2e2a56d2ab024ef50fe25742cddd65
+oid sha256:2641d9106fd9436f8698e9a0a7ca44fc88ef5a51aa6703df879379ad3af6aa8c
 size 4821600024

modelling_musilingo.py

@@ -2019,6 +2019,7 @@ class MusiLingo(BaseModel):
         self,
         mert_model,
         llama_model,
+        config,
         prompt_path="",
         prompt_template="",
         max_txt_len=32,
@@ -2259,6 +2260,7 @@ class MusilingoModel(PreTrainedModel):
         self.model = MusiLingo(
             mert_model=config.mert_model,
             llama_model=config.llama_model,
+            config=config,
             prompt_path=config.prompt_path,
             prompt_template=config.prompt_template,
             max_txt_len=config.max_txt_len,