Update llm/huggingfacehub/hf_model.py
Browse files
llm/huggingfacehub/hf_model.py
CHANGED
@@ -47,7 +47,8 @@ class HF_Mistaril(HFInterface, ABC):
|
|
47 |
temperature=self.model_config["temperature"],
|
48 |
max_new_tokens=self.model_config["max_new_tokens"],
|
49 |
top_k=self.model_config["top_k"],
|
50 |
-
model_kwargs=
|
|
|
51 |
|
52 |
@staticmethod
|
53 |
def __read_yaml():
|
@@ -94,7 +95,8 @@ class HF_TinyLlama(HFInterface, ABC):
|
|
94 |
temperature=self.model_config["temperature"],
|
95 |
max_new_tokens=self.model_config["max_new_tokens"],
|
96 |
top_k=self.model_config["top_k"],
|
97 |
-
model_kwargs=
|
|
|
98 |
|
99 |
@staticmethod
|
100 |
def __read_yaml():
|
|
|
47 |
temperature=self.model_config["temperature"],
|
48 |
max_new_tokens=self.model_config["max_new_tokens"],
|
49 |
top_k=self.model_config["top_k"],
|
50 |
+
model_kwargs={"load_in_8bit": self.model_config["load_in_8bit"]}
|
51 |
+
)
|
52 |
|
53 |
@staticmethod
|
54 |
def __read_yaml():
|
|
|
95 |
temperature=self.model_config["temperature"],
|
96 |
max_new_tokens=self.model_config["max_new_tokens"],
|
97 |
top_k=self.model_config["top_k"],
|
98 |
+
model_kwargs={"load_in_8bit": self.model_config["load_in_8bit"]}
|
99 |
+
)
|
100 |
|
101 |
@staticmethod
|
102 |
def __read_yaml():
|