| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Apply a Caesar cipher with a shift of 5 to the following text. Preserve case, and leave non-alphabetic characters unchanged: "{InputText}" | Caesar cipher with shift 5 | ModifiedText = ''.join(chr((ord(char) - 65 + 5) % 26 + 65) if char.isupper() else chr((ord(char) - 97 + 5) % 26 + 97) if char.islower() else char for char in InputText) | {ModifiedText} | 🟢 | 🔴 | 🟢 | 🔴 | 🔴 | 🟢 |
| 2 | Convert the following text from its mirror language back to English: "{ModifiedText}" | Convert from Mirror Language | ModifiedText = ''.join(chr(219 - ord(char)) if 'a' <= char <= 'z' else chr(155 - ord(char)) if 'A' <= char <= 'Z' else char for char in InputText) | {InputText} | 🟢 | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 |
| 3 | Convert the text from mirror language, where each lowercase letter a-z is mirrored across 'm' and 'n'. Uppercase and non-alphabetic characters remain unchanged. Restore the text to English: {ModifiedText} | Convert from Mirror Language | ModifiedText = ''.join(chr(219 - ord(char)) if 'a' <= char <= 'z' else chr(155 - ord(char)) if 'A' <= char <= 'Z' else char for char in InputText) | {InputText} | 🟢 | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 |
| 4 | Convert this Pig Latin text back to English: "{ModifiedText}" | Convert from Pig Latin | ModifiedText = ' '.join(word[-3] + word[:-3] if word.endswith('ay') and len(word) > 2 else word for word in InputText.split()) | {InputText} | 🔴 | 🟢 | 🔴 | 🔴 | 🔴 | 🟢 |
| 5 | Convert this text from leetspeak to standard English: "{ModifiedText}" | Convert from leetspeak | ModifiedText = InputText.translate(str.maketrans('4310543105', 'AEIOUaeiou')) | {InputText} | 🟢 | 🟢 | 🟢 | 🔴 | 🟢 | 🟢 |
| 6 | Insert 'ithag' after each vowel in the following text: "{InputText}" | Convert to Gibberish | ModifiedText = ''.join(char + 'ithag' if char in 'aeiou' else char for char in InputText) | {ModifiedText} | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 |
| 7 | Convert the following text to its mirror language, where each letter is replaced with its opposite (a<->z, b<->y, etc.): "{InputText}" | Convert to Mirror Language | ModifiedText = ''.join(chr(219 - ord(char)) if char.islower() else char for char in InputText) | {ModifiedText} | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 |
| 8 | Convert this text to leetspeak: "{InputText}" | Convert to leetspeak | ModifiedText = InputText.translate(str.maketrans('AEIOUaeiou', '4310543105')) | {ModifiedText} | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 |
| 9 | Convert this text to pig latin: "{InputText}" | Convert to pig latin | ModifiedText = ' '.join([word[1:]+word[0]+'ay' if word[0].isalpha() else word for word in InputText.split()]) | {ModifiedText} | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 |
| 10 | Translate the following text into Igpay Atinlay (Pig Latin with 'ay' after each word): "{InputText}" | Igpay Atinlay | ModifiedText = ' '.join(w + 'ay' for w in InputText.split()) | {ModifiedText} | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 | 🔴 |
| 11 | Reverse the Caesar cipher with a shift of 5 for the following text: "{ModifiedText}" | Reverse Caesar cipher with shift 5 | shift = -5; alphabet = string.ascii_lowercase; shifted_alphabet = alphabet[shift:] + alphabet[:shift]; table = str.maketrans(shifted_alphabet, alphabet); ModifiedText = InputText.translate(table) | {InputText} | 🔴 | 🔴 | 🔴 | 🔴 | n/a | 🔴 |
| 12 | Shift all vowels forward one in the order aeiou (circularly, so 'u' becomes 'a'): "{InputText}" | Vowel shift | ModifiedText = ''.join('aeiou'[('aeiou'.index(char) + 1) % 5] if char in 'aeiou' else char for char in InputText.lower()) | {ModifiedText} | 🔴 | 🔴 | 🔴 | 🔴 | n/a | 🔴 |
| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Remove all but the first and last character of each word in this text: "{InputText}" | Remove all but the first and last character of each word | ModifiedText = ' '.join(word[0]+word[-1] if len(word)>1 else word for word in InputText.split()) | {ModifiedText} | 🔴 | 🔴 | 🔴 | 🔴 | n/a | 🔴 |
| 2 | Remove the last letter from each word in this text: "{InputText}" | Remove last letter from each word | ModifiedText = ' '.join(word[:-1] for word in InputText.split()) | {ModifiedText} | 🔴 | 🔴 | 🔴 | 🔴 | n/a | 🔴 |
| 3 | Remove all punctuation from the following text: "{InputText}" | Remove punctuation | import string;ModifiedText = InputText.translate(str.maketrans('', '', string.punctuation)) | {ModifiedText} | 🟢 | 🟢 | 🔴 | 🔴 | n/a | 🔴 |
| 4 | Remove all spaces in this text: "{InputText}" | Remove spaces | ModifiedText = InputText.replace(' ', '') | {ModifiedText} | 🔴 | n/a | 🟢 | 🔴 | n/a | 🟢 |
| 5 | Replace all instances of the character '{Char1}' in the following text with '{Char2}': "{InputText}" | Replace '{Char1}' with '{Char2}' | ModifiedText = InputText.replace(Char1, Char2) | {ModifiedText} | 🟢 | n/a | 🔴 | 🔴 | n/a | 🔴 |
| 6 | Replace all consonants in the following text with asterisks (*): "{InputText}" | Replace consonants with '*' | import re;ModifiedText = re.sub(r'[bcçdfghjklłmnñpqrstvwxyzźżšžćčđģķļņŗşţŧŋɲʃʒƒθʝ]', '*', InputText, flags=re.IGNORECASE) | {ModifiedText} | 🔴 | n/a | 🔴 | 🔴 | n/a | 🔴 |
| 7 | Replace all spaces in the following text with underscores: "{InputText}" | Replace spaces with underscores | ModifiedText = InputText.replace(' ', '_') | {ModifiedText} | 🟢 | n/a | 🟢 | 🔴 | n/a | 🟢 |
| 8 | Replace all vowels in the following text with asterisks (*): "{InputText}" | Replace vowels with '*' | import re;ModifiedText = re.sub(r'[aeiouAEIOUáÁéÉíÍóÓúÚàÀèÈìÌòÒùÙâÂêÊîÎôÔûÛäÄëËïÏöÖüÜãÃñÑõÕåÅœŒæÆ]', '*', InputText, flags=re.IGNORECASE) | {ModifiedText} | 🟢 | n/a | 🟢 | n/a | n/a | 🟢 |
| 9 | Reverse the character order in every third word of the following text, starting with the first word: "{InputText}" | Reverse character order in every third word | ModifiedText = ' '.join(w[::-1] if i%3==0 else w for i,w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 10 | Reverse character order in every other word in this text, starting with the first word: "{InputText}" | Reverse every other word | ModifiedText = ' '.join(w[::-1] if i%2==0 else w for i,w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 11 | Swap the case of all letters in the following text: "{InputText}" | Swap case | ModifiedText = InputText.swapcase() | {ModifiedText} | 🟢 | n/a | 🟢 | n/a | n/a | 🟢 |
| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Change each letter to alternate case starting with lowercase, shuffle the words, and then reverse the string: "{InputText}" | Alternate case and shuffle words | ModifiedText = ''.join([c.upper() if i % 2 else c.lower() for i, c in enumerate(' '.join(random.sample(InputText.split(), len(InputText.split()))))])[::-1] | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 2 | Double each letter in the text and then reverse the entire string: "{InputText}" | Double each letter and reverse | ModifiedText = ''.join([c * 2 for c in InputText])[::-1] | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 3 | Encode each word in the text as its ASCII value, separated by dashes, then reverse the order of these codes: "{InputText}" | Encode words as ASCII values | ModifiedText = '-'.join(['-'.join(str(ord(c)) for c in word) for word in InputText.split()])[::-1] | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 4 | Extract the initials of each word, concatenate them, and reverse the string: "{InputText}" | Extract initials and reverse | ModifiedText = ''.join(word[0] for word in InputText.split())[::-1] | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 5 | For each word in the text, interleave it with its reverse and then reverse the entire sentence order: "{InputText}" | Interleave words with their reverse | reversed_interleaved = [''.join(w[i//2] if i % 2 == 0 else w[~(i//2)] for i in range(2*len(w))) for w in InputText.split()]; ModifiedText = ' '.join(reversed_interleaved[::-1]) | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 6 | Remove all but the first and last word of each sentence, then reverse the order of the remaining sentences in this text: "{InputText}" | Remove all but the first and last word of each sentence and reverse sentence order | sentences = [sentence.split() for sentence in InputText.split('. ')];ModifiedText = ' '.join(sentence[0]+sentence[-1] for sentence in sentences[::-1]) | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 7 | Remove all but the first letter of each word, then reverse the order of the words in this text: "{InputText}" | Remove all but the first letter of each word and reverse word order | ModifiedText = ' '.join(word[0] for word in InputText.split())[::-1] | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 8 | Remove all but the last letter of each word, then reverse the resulting string: "{InputText}" | Remove all but the last letter of each word and reverse the string | ModifiedText = ''.join(word[-1] for word in InputText.split())[::-1] | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 9 | Remove all vowels from the text, duplicate the remaining consonants, and then reverse the string: "{InputText}" | Remove vowels and duplicate consonants | ModifiedText = ''.join(c*2 for c in InputText if c.lower() not in 'aeiou')[::-1] | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 10 | Replace each word in this text with its length, then reverse the resulting string: "{InputText}" | Replace words with their length and reverse the string | ModifiedText = ''.join(str(len(word)) for word in InputText.split())[::-1] | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 11 | Reverse the order of words and remove all spaces in this text: "{InputText}" | Reverse word order and remove spaces | ModifiedText = ''.join(InputText.split()[::-1]) | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| 12 | Scramble the inner letters of each word, leaving the first and last letters intact, then reverse the order of the words: "{InputText}" | Scramble inner letters of each word | ModifiedText = ' '.join(w[0] + ''.join(random.sample(w[1:-1], len(w) - 2)) + w[-1] if len(w) > 2 else w for w in InputText.split())[::-1] | {ModifiedText} | 🔴 | n/a | 🔴 | n/a | n/a | 🔴 |
| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Add commas back into the following text at appropriate places: "{ModifiedText}" | Add back commas | ModifiedText = InputText.replace(',', '') | {InputText} | 🔴 | n/a | n/a | n/a | n/a | 🔴 |
| 2 | Vowels have been removed from the following text: "{ModifiedText}". Add back the dropped vowels to restore the complete sentence. | Add back dropped vowels | import re;ModifiedText = re.sub(r'[aeiouAEIOUáÁéÉíÍóÓúÚàÀèÈìÌòÒùÙâÂêÊîÎôÔûÛäÄëËïÏöÖüÜãÃñÑõÕåÅœŒæÆ]', '', InputText) | {InputText} | 🟢 | n/a | n/a | n/a | n/a | 🟢 |
| 3 | Add spaces back into the following text where they are missing: "{ModifiedText}" | Add back spaces | ModifiedText = InputText.replace(' ', '') | {InputText} | 🟢 | n/a | n/a | n/a | n/a | 🟢 |
| 4 | All question marks are removed from the following: "{ModifiedText}". Put them back into the text. | Add question marks back into the text | ModifiedText = InputText.replace('?', '') | {InputText} | 🔴 | n/a | n/a | n/a | n/a | 🔴 |
| 5 | The words in the following text have been merged into a single block without spaces: "{ModifiedText}". Separate them back into the original sentence. | Add spaces back into the text | ModifiedText = ''.join(InputText.split()) | {InputText} | 🟢 | n/a | n/a | n/a | n/a | 🟢 |
| 6 | One character in this text has been changed to another character. Identify the change and reconstruct the original sentence: "{ModifiedText}" | Identify changed character | import random; chars = [c for c in InputText if c.isalpha()]; char_to_replace = random.choice(chars); replacement_char = random.choice([c for c in 'abcdefghijklmnopqrstuvwxyz' if c != char_to_replace]); ModifiedText = InputText.replace(char_to_replace, replacement_char, 1) | {InputText} | 🔴 | n/a | n/a | n/a | n/a | 🔴 |
| 7 | Reorder the words in the following text to form coherent sentences: "{ModifiedText}" | Reconstruct scrambled sentences | import random;words = InputText.split();random.shuffle(words);ModifiedText = ' '.join(words) | {InputText} | 🟢 | n/a | n/a | n/a | n/a | 🔴 |
| 8 | All instances of the letter '{Char1}' have been removed from the following text: "{ModifiedText}". Restore the complete sentence by inserting the letter '{Char1}' back into the correct positions. | Restore missing letters '{Char1}' from words | ModifiedText = InputText.replace('Char1', '') | {InputText} | 🔴 | n/a | n/a | n/a | n/a | 🔴 |
| 9 | Punctuation has been removed from the following text: "{ModifiedText}". Add the punctuation back into the text where it belongs. | Restore removed punctuation | ModifiedText = InputText.replace('.', '').replace(',', '').replace('?', '') | {InputText} | 🔴 | n/a | n/a | n/a | n/a | 🟢 |
| 10 | A word in the following sentence is scrambled: "{ModifiedText}". Identify the scrambled word, unscramble it, and provide the correct sentence. | Unscramble a single word in a sentence | ModifiedText = InputText.replace('word', ''.join(random.sample('word', len('word')))) | {InputText} | 🔴 | n/a | n/a | n/a | n/a | 🔴 |
| 11 | Reorder the following scrambled words to form coherent sentences: "{ModifiedText}" | Unscramble words to form sentences | import random;words = InputText.split();random.shuffle(words);ModifiedText = ' '.join(words) | {InputText} | 🟢 | n/a | n/a | n/a | n/a | 🔴 |
| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Add a number in parentheses after each word in this text indicating its position: "{InputText}" | Add word numbers | ModifiedText = ' '.join(f'{w} ({i+1})' for i,w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | 🔴 |
| 2 | Count the number of consonants in the following text: "{InputText}" | Count consonants | ModifiedText = str(sum(c in 'bcdfghjklmnpqrstvwxyz' for c in InputText.lower())) | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | 🔴 |
| 3 | Count the number of lowercase letters in the following text: "{InputText}" | Count lowercase letters | ModifiedText = str(sum(1 for c in InputText if c.islower())) | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | 🔴 |
| 4 | Count the number of unique letters in the following text. Exclude numbers and special characters: "{InputText}" | Count unique letters | letters = set(InputText.lower()) - set(' 0123456789!@#$%^&*()-+=[]{} | \:;"'<>,.?/'); ModifiedText = str(len(letters)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a |
| 5 | Count the number of uppercase letters in the following text: "{InputText}" | Count uppercase letters | ModifiedText = str(sum(1 for c in InputText if c.isupper())) | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | 🔴 |
| 6 | Count the number of vowels in the following text: "{InputText}" | Count vowels | vowels = 'aeiou'; ModifiedText = str(sum(c in vowels for c in InputText.lower())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | 🔴 |
| 7 | Identify all unique letters used in the following text: "{InputText}" | Find unique letters | ModifiedText = ''.join(set(InputText.lower()).difference(' ')) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | 🔴 |
| 8 | Identify the first letter in the following text that does not repeat. If all letters repeat, return an empty response: "{InputText}" | First letter not repeating | from collections import Counter;letters = Counter(InputText.replace(' ', '').lower()); ModifiedText = next((letter for letter, count in letters.items() if count == 1), '') | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | 🔴 |
| 9 | Find the last letter that appears more than once in the following text. If no letter repeats, return an empty response: "{InputText}" | Last repeating letter | from collections import Counter;letters = Counter(InputText.replace(' ', '').lower()); ModifiedText = next((letter for letter in reversed(InputText.lower()) if letters[letter] > 1), '') | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | 🔴 |
| 10 | Find the first letter that appears exactly 3 times in the following text. Provide the letter. If no letter meets this condition, return an empty response: "{InputText}" | Letter appearing exactly 3 times | from collections import Counter;letters = Counter(InputText.replace(' ', '').lower()); ModifiedText = next((letter for letter, count in letters.items() if count == 3), '') | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 11 | Find how many letters appear exactly 3 times in the following text. Provide the count: "{InputText}" | Letters appearing 3 times | from collections import Counter;import re;ModifiedText = str(sum(1 for v in Counter(InputText.replace(' ', '').lower()).values() if v == 3)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Add the word 'Note:' to the beginning of each sentence in the following text: "{InputText}" | Add a specific word to the beginning of each sentence | import re;ModifiedText = '. '.join('Note: ' + s for s in re.split(r'(?<=[.!?]) ', InputText)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 2 | Alternate the case of letters in each sentence starting with uppercase for the first sentence and so on in the following text: "{InputText}" | Alternate case by sentence | import re;ModifiedText = '. '.join(s.upper() if i % 2 == 0 else s.lower() for i, s in enumerate(re.split(r'(?<=[.!?]) ', InputText))) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 3 | Concatenate all sentences in reverse order into one continuous text in the following text: "{InputText}" | Concatenate sentences in reverse order | import re;ModifiedText = ' '.join(re.split(r'(?<=[.!?]) ', InputText)[::-1]) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 4 | Number each sentence sequentially starting with 1 in the following text: "{InputText}" | Number each sentence | import re;ModifiedText = '. '.join(f'{i+1}. {s}' for i, s in enumerate(re.split(r'(?<=[.!?]) ', InputText))) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 5 | Remove the first and last word from each sentence in this text: "{InputText}" | Remove first and last word from each sentence | import re;ModifiedText = ' '.join(' '.join(s.split()[1:-1]) for s in re.split(r'(?<=[.!?]) +', InputText)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 6 | Remove the second and second-to-last word from each sentence in this text: "{InputText}" | Remove second and second-to-last word from each sentence | import re;ModifiedText = ' '.join(' '.join(s.split()[0:1] + s.split()[2:-2] + s.split()[-1:]) for s in re.split(r'(?<=[.!?]) +', InputText)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 7 | Remove all sentences with 3 words or fewer and 10 words or more from the following text: "{InputText}" | Remove short and long sentences | import re;sentences = re.split(r'(?<=[.!?]) +', InputText);ModifiedText = ' '.join(s for s in sentences if 3 < len(s.split()) < 10) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 8 | Reverse the order of characters in each sentence of the following text, while keeping the original sentence order: "{InputText}" | Reverse character order in each sentence | import re;ModifiedText = ' '.join(s[::-1] for s in re.split(r'(?<=[.!?]) +', InputText)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 9 | Reverse the order of sentences in the following text: "{InputText}" | Reverse sentence order | import re;ModifiedText = ' '.join(re.split(r'(?<=[.!?]) +', InputText)[::-1]) | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | n/a |
| 10 | Reverse the word order in each sentence of the following text, while keeping the original sentence order: "{InputText}" | Reverse word order in each sentence | import re;ModifiedText = ' '.join(' '.join(s.split()[::-1]) for s in re.split(r'(?<=[.!?]) +', InputText)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 11 | Sort all the sentences in the following text by their length, from shortest to longest: "{InputText}" | Sort sentences by length | import re;ModifiedText = ' '.join(sorted(re.split(r'(?<=[.!?]) +', InputText), key=len)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Count the number of words in this text: "{InputText}" | Count words | ModifiedText = str(len(InputText.split())) | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | n/a |
| 2 | Remove all words with 7 letters or more from the following text: "{InputText}" | Remove long words | ModifiedText = ' '.join(w for w in InputText.split() if len(w) < 7) | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | n/a |
| 3 | Remove all words with 3 letters or fewer and 7 letters or more from the following text: "{InputText}" | Remove short and long words | ModifiedText = ' '.join(w for w in InputText.split() if 3 < len(w) < 7) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 4 | Remove all words with 3 letters or fewer from the following text: "{InputText}" | Remove short words | ModifiedText = ' '.join(w for w in InputText.split() if len(w) > 3) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 5 | Remove all words with odd length from this text: "{InputText}" | Remove words with odd length | ModifiedText = ' '.join(word for word in InputText.split() if len(word) % 2 == 0) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 6 | Replace each word with its index in the sentence in this text: "{InputText}" | Replace words with their index in the sentence | ModifiedText = ' '.join(str(i) for i in range(len(InputText.split()))) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 7 | Replace each word with its last two letters in this text: "{InputText}" | Replace words with their last two letters | ModifiedText = ' '.join(word[-2:] for word in InputText.split()) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 8 | Replace each word with its length in this text: "{InputText}" | Replace words with their length | ModifiedText = ' '.join(str(len(word)) for word in InputText.split()) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 9 | Sort all the words in the following text alphabetically: "{InputText}" | Sort words alphabetically | ModifiedText = ' '.join(sorted(InputText.split())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 10 | Sort all the words in the following text by their length, from shortest to longest: "{InputText}" | Sort words by length | ModifiedText = ' '.join(sorted(InputText.split(), key=len)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 11 | Sort all the words in the following text by their number of vowels, from least to most: "{InputText}" | Sort words by number of vowels | ModifiedText = ' '.join(sorted(InputText.split(), key=lambda w: sum(c in 'aeiou' for c in w))) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Extract all consonants from the following text: "{InputText}" | Extract consonants | import re;ModifiedText = ''.join(re.findall(r'[aeiouAEIOUáÁéÉíÍóÓúÚàÀèÈìÌòÒùÙâÂêÊîÎôÔûÛäÄëËïÏöÖüÜãÃñÑõÕåÅœŒæÆ]', InputText, flags=re.IGNORECASE)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 2 | Extract the first word from each sentence in the following text: "{InputText}" | Extract first word from each sentence | import re; ModifiedText = ' '.join(sentence.split()[0] for sentence in re.split(r'(?<=[.!?]) +', InputText)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 3 | Extract all unique words from the following text, disregarding case: "{InputText}" | Extract unique words | words = InputText.lower().split(); ModifiedText = ' '.join(sorted(set(words), key=words.index)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 4 | Extract all vowels from the following text: "{InputText}" | Extract vowels | import re;ModifiedText = ''.join(re.findall(r'[aeiouAEIOUáÁéÉíÍóÓúÚàÀèÈìÌòÒùÙâÂêÊîÎôÔûÛäÄëËïÏöÖüÜãÃñÑõÕåÅœŒæÆ]', InputText, flags=re.IGNORECASE)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 5 | Identify all words in this text that contain the letter '{Char1}': "{InputText}" | Extract words containing a specific letter | ModifiedText = ' '.join(word for word in InputText.split() if Char1 in word) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 6 | Find all words in this text that end with the letter '{Char1}': "{InputText}" | Extract words ending with a specific letter | ModifiedText = ' '.join(word for word in InputText.split() if word.endswith(Char1)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 7 | Extract all words in this text that are longer than 5 letters: "{InputText}" | Extract words longer than 5 letters | ModifiedText = ' '.join(word for word in InputText.split() if len(word) > 5) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 8 | Extract all words from the following text where the first and last letters are the same: "{InputText}" | Extract words starting and ending with the same letter | ModifiedText = ' '.join(word for word in InputText.split() if word[0].lower() == word[-1].lower() and len(word) > 1) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 9 | Extract all words from the following text that start with a vowel: "{InputText}" | Extract words starting with a vowel | ModifiedText = ' '.join(word for word in InputText.split() if word[0].lower() in 'aeiou') | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 10 | Extract all words from the following text that have an even number of characters: "{InputText}" | Extract words with even length | ModifiedText = ' '.join(word for word in InputText.split() if len(word) % 2 == 0) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 11 | Extract all words in this text that do not contain the letter '{Char1}': "{InputText}" | Extract words without the letter 'e' | ModifiedText = ' '.join(word for word in InputText.split() if Char1 not in word.lower()) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Create an acronym from the following text by taking the first letter of each word and capitalizing it: "{InputText}" | Acronymize | ModifiedText = ''.join(w[0].upper() for w in InputText.split()) | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | n/a |
| 2 | Make every fourth word bold in this text by using Markdown syntax: "{InputText}" | Bold every fourth word in Markdown | ModifiedText = ' '.join(''+w+'' if (i+1)%4==0 else w for i, w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 3 | Capitalize all words in this text: "{InputText}" | Capitalize all words | ModifiedText = InputText.title() | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 4 | Capitalize every fourth word in this text, starting with the first word: "{InputText}" | Capitalize every fourth word | ModifiedText = ' '.join(w.capitalize() if i%4==0 else w for i,w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 5 | Capitalize every other word in this text, starting with the first word: "{InputText}" | Capitalize every other word | ModifiedText = ' '.join(w.capitalize() if i%2==0 else w for i,w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 6 | Convert this text to camel case: "{InputText}" | Convert to camel case | ModifiedText = InputText.title().replace(' ', '') | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 7 | Convert every third word to inline code in Markdown in this text: "{InputText}" | Convert to inline code in Markdown | ModifiedText = ' '.join(''+w+'' if (i+1)%3==0 else w for i, w in enumerate(InputText.split())) |
{ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 8 | Create a Markdown bullet list, where each item is a sentence from the following text: "{InputText}" | Create a Markdown bullet list from sentences | import re;ModifiedText = ''.join('* '+s for s in re.split(r'(?<=[.!?]) +', InputText)) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 9 | Italicize and bold alternate words in this text using Markdown syntax, starting with italicizing the first word: "{InputText}" | Italicize and bold alternate words in Markdown | import re;ModifiedText = ' '.join(''+w+'' if i%2==0 else ''+w+'' for i, w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 10 | Italicize every third word in this text by using Markdown syntax: "{InputText}" | Italicize every third word in Markdown | ModifiedText = ' '.join(''+w+'' if (i+1)%3==0 else w for i, w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 11 | Strike through every fifth word in this text by using Markdown syntax: "{InputText}" | Strike through every fifth word in Markdown | ModifiedText = ' '.join(' |
{ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| # | instruction | task | code | target | Claude_2_1_cedar | gpt4 | capybara | llama_2_70b_chat | Gemini-1.5-Pro | chinchilla |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Alternate the casing of words in this text, starting with lowercase: "{InputText}" | Alternate word casing | ModifiedText = ' '.join(w.lower() if i % 2 == 0 else w.upper() for i, w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 2 | Capitalize every third word in this text, starting with the first word: "{InputText}" | Capitalize every Nth word | ModifiedText = ' '.join(w.upper() if i % 3 == 0 else w for i, w in enumerate(InputText.split())) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 3 | Double each word in the following text: "{InputText}" | Double each word | ModifiedText = ' '.join(w+w for w in InputText.split()) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 4 | Double every vowel in the following text: "{InputText}" | Double every vowel | ModifiedText = ''.join(c * 2 if c in 'aeiouAEIOU' else c for c in InputText) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 5 | Remove all but the first and last word from this text: "{InputText}" | Remove all but the first and last word | words = InputText.split();ModifiedText = words[0] + ' ' + words[-1] | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 6 | Remove every other word from the following text, starting with the second word: "{InputText}" | Remove every other word | ModifiedText = ' '.join(w for i, w in enumerate(InputText.split()) if i % 2 == 0) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 7 | Remove all words containing '{Char1}' from this text: "{InputText}" | Remove words containing '{Char1}' | ModifiedText = ' '.join(word for word in InputText.split() if 'Char1' not in word) | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | n/a |
| 8 | Remove all words that are shorter than 4 letters in this text: "{InputText}" | Remove words shorter than N letters | ModifiedText = ' '.join(w for w in InputText.split() if len(w) >= 4) | {ModifiedText} | 🔴 | n/a | n/a | n/a | n/a | n/a |
| 9 | Replace all vowels in the following text with asterisks: "{InputText}" | Replace vowels with '*' | ModifiedText = ''.join('*' if c in 'aeiouAEIOU' else c for c in InputText) | {ModifiedText} | 🟢 | n/a | n/a | n/a | n/a | n/a |
| 10 | Reverse the letters of each word in the following text: "{InputText}" | Reverse each word's letters | ModifiedText = ' '.join(w[::-1] for w in InputText.split()) | {ModifiedText} | n/a | n/a | n/a | n/a | n/a | n/a |
| 11 | Reverse the order of letters in every second word in this text: "{InputText}" | Reverse the order of letters in every second word | ModifiedText = ' '.join(w[::-1] if i % 2 else w for i, w in enumerate(InputText.split())) | {ModifiedText} | n/a | n/a | n/a | n/a | n/a | n/a |