realPCH commited on
Commit
041a246
1 Parent(s): 4951e82

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +26 -0
README.md ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: mit
3
+ datasets:
4
+ - kyujinpy/KOR-gugugu-platypus-set
5
+ ---
6
+ ### Developed by chPark
7
+
8
+ ### Training Strategy
9
+ We fine-tuned this model based on [yanolja/KoSOLAR-10.7B-v0.1](https://huggingface.co/yanolja/KoSOLAR-10.7B-v0.1-deprecated) with [kyujinpy/KOR-gugugu-platypus-set](https://huggingface.co/datasets/kyujinpy/KOR-gugugu-platypus-set)
10
+
11
+ ### Run the model
12
+ ```python
13
+ from transformers import AutoModelForCausalLM, AutoTokenizer
14
+
15
+ model_id = "realPCH/ko_solra_merge"
16
+ tokenizer = AutoTokenizer.from_pretrained(model_id)
17
+
18
+ model = AutoModelForCausalLM.from_pretrained(model_id)
19
+
20
+ text = "[INST] Put instruction here. [/INST]"
21
+ inputs = tokenizer(text, return_tensors="pt")
22
+
23
+ outputs = model.generate(**inputs, max_new_tokens=20)
24
+ print(tokenizer.decode(outputs[0], skip_special_tokens=True))
25
+
26
+ ```