Update README.md
Browse files
README.md
CHANGED
|
@@ -21,6 +21,7 @@ An experimental Twitter (**X**) bot is available at [https://twitter.com/ZenosBo
|
|
| 21 |
The latest development version of Transformers, which includes serialization of 4 bits models.
|
| 22 |
|
| 23 |
- [Transformers](https://huggingface.co/docs/transformers/installation#install-from-source)
|
|
|
|
| 24 |
|
| 25 |
Since this is a compressed version (4 bits), it can fit into ~7GB of VRAM.
|
| 26 |
|
|
@@ -33,7 +34,7 @@ from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig
|
|
| 33 |
|
| 34 |
# Load the tokenizer and model
|
| 35 |
tokenizer = AutoTokenizer.from_pretrained("webpolis/zenos-gpt-j-6B-instruct-4bit")
|
| 36 |
-
model = AutoModelForCausalLM.from_pretrained("webpolis/zenos-gpt-j-6B-instruct-4bit")
|
| 37 |
|
| 38 |
user_msg = '''Escribe un poema breve utilizando los siguientes conceptos:
|
| 39 |
|
|
|
|
| 21 |
The latest development version of Transformers, which includes serialization of 4 bits models.
|
| 22 |
|
| 23 |
- [Transformers](https://huggingface.co/docs/transformers/installation#install-from-source)
|
| 24 |
+
- Bitsandbytes >= 0.41.3
|
| 25 |
|
| 26 |
Since this is a compressed version (4 bits), it can fit into ~7GB of VRAM.
|
| 27 |
|
|
|
|
| 34 |
|
| 35 |
# Load the tokenizer and model
|
| 36 |
tokenizer = AutoTokenizer.from_pretrained("webpolis/zenos-gpt-j-6B-instruct-4bit")
|
| 37 |
+
model = AutoModelForCausalLM.from_pretrained("webpolis/zenos-gpt-j-6B-instruct-4bit", use_safetensors=True)
|
| 38 |
|
| 39 |
user_msg = '''Escribe un poema breve utilizando los siguientes conceptos:
|
| 40 |
|