PeterZentai
commited on
Commit
•
cb64c73
1
Parent(s):
a11eb6d
Update README.md
Browse files
README.md
CHANGED
@@ -45,6 +45,8 @@ Cost for inference.
|
|
45 |
```python
|
46 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
47 |
|
|
|
|
|
48 |
model = AutoModelForCausalLM.from_pretrained(model_id, device_map={"": 0}, trust_remote_code=True)
|
49 |
tokenizer = tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=True, trust_remote_code=True)
|
50 |
|
@@ -63,4 +65,7 @@ with torch.no_grad():
|
|
63 |
# but it might continue generating irrelevant text. this way the model will stop at the right place
|
64 |
model_response = model.generate(**model_input, max_new_tokens=512, eos_token_id=tokenizer.eos_token_id, )
|
65 |
print(tokenizer.decode(model_result[0], skip_special_tokens=False))
|
66 |
-
```
|
|
|
|
|
|
|
|
45 |
```python
|
46 |
from transformers import AutoTokenizer, AutoModelForCausalLM
|
47 |
|
48 |
+
model_id = "WeeRobots/phi-2-chat-v05"
|
49 |
+
|
50 |
model = AutoModelForCausalLM.from_pretrained(model_id, device_map={"": 0}, trust_remote_code=True)
|
51 |
tokenizer = tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=True, trust_remote_code=True)
|
52 |
|
|
|
65 |
# but it might continue generating irrelevant text. this way the model will stop at the right place
|
66 |
model_response = model.generate(**model_input, max_new_tokens=512, eos_token_id=tokenizer.eos_token_id, )
|
67 |
print(tokenizer.decode(model_result[0], skip_special_tokens=False))
|
68 |
+
```
|
69 |
+
|
70 |
+
# Non production quality
|
71 |
+
Be aware that this model tuning wasn't thoroughly tested, and isn't meant to be used in production, only for experimentation or hobby projects.
|