Update app.py
Browse files
app.py
CHANGED
|
@@ -82,7 +82,6 @@ model = AutoModelForCausalLM.from_pretrained(model_id, token= token,
|
|
| 82 |
# torch_dtype=torch.fl,
|
| 83 |
# attn_implementation="flash_attention_2",
|
| 84 |
# low_cpu_mem_usage=True,
|
| 85 |
-
quantization_config=bnb.QuantizationConfig(bits=8),
|
| 86 |
# device_map='cuda',
|
| 87 |
# device_map=accelerator.device_map,
|
| 88 |
|
|
|
|
| 82 |
# torch_dtype=torch.fl,
|
| 83 |
# attn_implementation="flash_attention_2",
|
| 84 |
# low_cpu_mem_usage=True,
|
|
|
|
| 85 |
# device_map='cuda',
|
| 86 |
# device_map=accelerator.device_map,
|
| 87 |
|