FPHam commited on
Commit
c8d3c9c
1 Parent(s): 668b7b2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -6
README.md CHANGED
@@ -21,11 +21,10 @@ Ah, Karen, a true peach among grammatical cucumbers! She yearns to rectify the m
21
  Yet, unlike those ChatGPT kaboodles that morph into self-absorbed, constipated gurus of self-help style, Karen remains steadfastly grounded in grammatical wisdom but respectfull of your style.
22
 
23
  # Info
24
- Karen V2 uses completely different dataset and base model than the previous Karen
25
 
26
  # Goals
27
- Karen's goals are fixing grammar and spelling errors with as lite changes to the style as possible.
28
- It's tuned to catch most typical ESL errors.
29
 
30
  Verb Tense Errors:
31
  Incorrect use of verb tenses, such as using present tense when past tense is required and vice versa.
@@ -65,10 +64,10 @@ It's tuned to catch most typical ESL errors.
65
  Mistakes in matching singular and plural nouns and verbs in a sentence.
66
 
67
  # Future Goals
68
- Add more grammar error cases, better dataset, use larger dataset
69
 
70
  # Training
71
- It was reverse trained on fiction paragraphs where errors were deliberately introduced by another LLama model and python script.
72
 
73
  # Usage
74
  It should be used by submitting a paragraph or block of text at a time.
@@ -82,7 +81,7 @@ It should be used by submitting a paragraph or block of text at a time.
82
  Edit the following text for spelling and grammar mistakes: {paragraph of text} <|im_end|>
83
  <|im_start|>assistant
84
  ```
85
- Note the pretext: *Edit the following text for spelling and grammar mistakes:* before the actual text. It ensures that Karen won't start talking *about* the text.
86
 
87
  # Recomended settings
88
 
 
21
  Yet, unlike those ChatGPT kaboodles that morph into self-absorbed, constipated gurus of self-help style, Karen remains steadfastly grounded in grammatical wisdom but respectfull of your style.
22
 
23
  # Info
24
+ Karen, Version 2, uses a completely different data set and base model than the previous Karen.
25
 
26
  # Goals
27
+ Karen's goal is to fix grammatical and spelling errors with as light a change to the style as possible. It's tuned to catch most typical ESL errors.
 
28
 
29
  Verb Tense Errors:
30
  Incorrect use of verb tenses, such as using present tense when past tense is required and vice versa.
 
64
  Mistakes in matching singular and plural nouns and verbs in a sentence.
65
 
66
  # Future Goals
67
+ Add more grammar error cases. Better datasets. Use larger datasets.
68
 
69
  # Training
70
+ It was reversely trained on fictional passages where errors were intentionally inserted by another Llama model and Python script.
71
 
72
  # Usage
73
  It should be used by submitting a paragraph or block of text at a time.
 
81
  Edit the following text for spelling and grammar mistakes: {paragraph of text} <|im_end|>
82
  <|im_start|>assistant
83
  ```
84
+ Note the pretext: *Edit the following text for spelling and grammar mistakes:* before the actual text. This way Karen wouldn't start talking ABOUT the text.
85
 
86
  # Recomended settings
87