Update README.md
Browse files
README.md
CHANGED
@@ -8,7 +8,7 @@ tags:
|
|
8 |
- solar
|
9 |
license: apache-2.0
|
10 |
---
|
11 |
-
#
|
12 |
|
13 |
This is a SOLAR-like model upscaled to 18B.
|
14 |
It is a frankenmerge model created using mergekit, alternating layers of Nous-Hermes-2-SOLAR-10.7B and SOLAR-10.7B-Instruct.
|
@@ -69,8 +69,8 @@ dtype: float16
|
|
69 |
You can use the provided template:
|
70 |
|
71 |
```
|
72 |
-
tokenizer = AutoTokenizer.from_pretrained("vicgalle/
|
73 |
-
model = AutoModelForCausalLM.from_pretrained("vicgalle/
|
74 |
|
75 |
conversation = [ {'role': 'system', 'content': SYSTEM_PROMPT}, {'role': 'user', 'content': USER_PROMPT} ]
|
76 |
prompt = tokenizer.apply_chat_template(conversation, tokenize=False, add_generation_prompt=True)
|
|
|
8 |
- solar
|
9 |
license: apache-2.0
|
10 |
---
|
11 |
+
# vicgalle/franken-SOLAR-18B-v1.0
|
12 |
|
13 |
This is a SOLAR-like model upscaled to 18B.
|
14 |
It is a frankenmerge model created using mergekit, alternating layers of Nous-Hermes-2-SOLAR-10.7B and SOLAR-10.7B-Instruct.
|
|
|
69 |
You can use the provided template:
|
70 |
|
71 |
```
|
72 |
+
tokenizer = AutoTokenizer.from_pretrained("vicgalle/franken-SOLAR-18B-v1.0")
|
73 |
+
model = AutoModelForCausalLM.from_pretrained("vicgalle/franken-SOLAR-18B-v1.0", torch_dtype=torch.float16, load_in_4bit=True)
|
74 |
|
75 |
conversation = [ {'role': 'system', 'content': SYSTEM_PROMPT}, {'role': 'user', 'content': USER_PROMPT} ]
|
76 |
prompt = tokenizer.apply_chat_template(conversation, tokenize=False, add_generation_prompt=True)
|