Spaces:
Running
Running
Upload app.py with huggingface_hub
Browse files
app.py
CHANGED
|
@@ -217,7 +217,7 @@ def load_model(model_key: str):
|
|
| 217 |
model = Llama.from_pretrained(
|
| 218 |
repo_id=config['repo_id'],
|
| 219 |
filename=config['filename'],
|
| 220 |
-
n_ctx=
|
| 221 |
n_threads=8,
|
| 222 |
verbose=False,
|
| 223 |
)
|
|
@@ -234,7 +234,7 @@ def preload_all_models():
|
|
| 234 |
_loaded_models[key] = Llama.from_pretrained(
|
| 235 |
repo_id=config['repo_id'],
|
| 236 |
filename=config['filename'],
|
| 237 |
-
n_ctx=
|
| 238 |
n_threads=8,
|
| 239 |
verbose=False,
|
| 240 |
)
|
|
|
|
| 217 |
model = Llama.from_pretrained(
|
| 218 |
repo_id=config['repo_id'],
|
| 219 |
filename=config['filename'],
|
| 220 |
+
n_ctx=512,
|
| 221 |
n_threads=8,
|
| 222 |
verbose=False,
|
| 223 |
)
|
|
|
|
| 234 |
_loaded_models[key] = Llama.from_pretrained(
|
| 235 |
repo_id=config['repo_id'],
|
| 236 |
filename=config['filename'],
|
| 237 |
+
n_ctx=512,
|
| 238 |
n_threads=8,
|
| 239 |
verbose=False,
|
| 240 |
)
|