Corianas commited on
Commit
67e7541
·
verified ·
1 Parent(s): 73eea83

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +38 -0
README.md ADDED
@@ -0,0 +1,38 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-nc-sa-4.0
3
+ datasets:
4
+ - roneneldan/TinyStories
5
+ - guanaco/guanaco_clean
6
+ ---
7
+ This is a character (english a-z 0-9 and so on) trained model following Andrej karpathy's llama.c project https://github.com/karpathy/llama2.c on both TinyStories and my own internal similar dataset I made.
8
+
9
+ It has been finetuned on a question answer set, without any preamble, you ask the question, give a newline, and wait for the answer.
10
+
11
+ for it to see/output Uppercase letters this model uses a Shift-Key modifier before the letter to become uppercase, and has never been trained on actual uppercase letters.
12
+
13
+ This modifier is ↨ and here are the functions I use to convert from straight text to the modified format and back.
14
+ ```
15
+ def add_caseifer(text):
16
+ # Using list comprehension for more efficient concatenation
17
+ return ''.join(['↨' + char.lower() if char.isupper() else char for char in text
18
+
19
+ def remove_caseifer(text):
20
+ new_text = ""
21
+ i = 0
22
+ while i < len(text):
23
+ if text[i] == "↨":
24
+ if i+1 < len(text):
25
+ new_text += text[i+1].upper()
26
+ i += 1
27
+ else:
28
+ pass # skip this index
29
+ else:
30
+ new_text += text[i]
31
+ i += 1
32
+ return new_text
33
+ ```
34
+
35
+ As such for test strings to use in chat try using somthing like:
36
+ ```
37
+ ↨hello, my name is ↨clara and ↨i like
38
+ ```