Update README.md
Browse files
README.md
CHANGED
|
@@ -95,7 +95,7 @@ model = AutoModelForCausalLM.from_pretrained(
|
|
| 95 |
model_id,
|
| 96 |
torch_dtype=torch.bfloat16,
|
| 97 |
device_map="auto",
|
| 98 |
-
)
|
| 99 |
|
| 100 |
messages = [...] # add message here
|
| 101 |
|
|
|
|
| 95 |
model_id,
|
| 96 |
torch_dtype=torch.bfloat16,
|
| 97 |
device_map="auto",
|
| 98 |
+
) # We don't recommend using BNB 4-bit (load_in_4bit) here. Instead, use AWQ, as detailed here: https://huggingface.co/scb10x/llama-3-typhoon-v1.5x-70b-instruct-awq.
|
| 99 |
|
| 100 |
messages = [...] # add message here
|
| 101 |
|