Datasets:

kipasyangin5
/

5m-terminal-lm-dataset

text stringlengths 0 3.69k
Question: How do I stage all changed files in git?
Answer:
```bash
git add .
```

Question: How do I monitor a log file in real-time?
Answer:
```bash
tail -f /var/log/syslog
```

Question: How do I monitor a log file in real-time?
Answer:
```bash
tail -f /var/log/syslog
```

Question: How do I check repository status in git?
Answer:
```bash
git status
```

Question: How do I check disk usage of current directories?
Answer:
```bash
du -sh * \| sort -hr
```

Question: How do I list files with human readable file sizes?
Answer:
```bash
ls -lh
```

User: How do I navigate up one directory?
Assistant: Run `cd ..`

Question: How do I copy a directory recursively?
Answer:
```bash
cp -r source_dir/ target_dir/
```

User: How do I make a shell script executable?
Assistant: Run `chmod +x script.sh`

Question: How do I go to the home directory?
Answer:
```bash
cd ~
```

$ head -n 20 logfile.log
# Executed successfully

User: How do I stage all changed files in git?
Assistant: Run `git add .`

Question: How do I check my current directory path?
Answer:
```bash
pwd
```

$ wc -l dataset.txt
# Executed successfully

Question: How do I navigate up one directory?
Answer:
```bash
cd ..
```

User: How do I create an empty file?
Assistant: Run `touch index.js`

$ rm -rf target_folder/
# Executed successfully

Question: How do I monitor a log file in real-time?
Answer:
```bash
tail -f /var/log/syslog
```

Question: How do I push commits to remote main branch?
Answer:
```bash
git push origin main
```

User: How do I recursively search for text in files?
Assistant: Run `grep -rn "search_term" .`

Question: How do I list files with human readable file sizes?
Answer:
```bash
ls -lh

End of preview. Expand in Data Studio

📦 50 Million Token Terminal & Multilingual Corpus

This dataset contains the exact 51,609,269 Token raw text corpus (100.54 MB uncompressed / 32.46 MB compressed) used to train the 5M Terminal & Multilingual Language Models (kipasyangin5/5m-terminal-lm-chinchilla and kipasyangin5/5m-terminal-lm-saturated).

Dataset Composition

Total Tokens: 51,609,269 Tokens
Raw Uncompressed Size: 100.54 MB (dataset_50m.txt)
Compressed Gzip Size: 32.46 MB (dataset_50m.txt.gz)

Mixture Breakdown:

Linux Terminal & CLI Commands (20%): Linux navigation (cd .., pwd, ls -la), searching (grep, find), permissions (chmod), Git operations, networking (curl), package managers, Docker, and Q&A pairs.
English General Language (24%): wikitext-2-raw-v1 / wikitext-103 articles & tech prose.
Multilingual Language (56%): Multilingual text spanning Indonesian, Spanish, French, German.

Files Included

dataset_50m.txt.gz: The full compressed 50M token raw text stream.
dataset_sample.txt: A 1,000 line raw text sample for instant browser preview.

Usage in Python

import gzip

# Decompress and read dataset
with gzip.open("dataset_50m.txt.gz", "rt", encoding="utf-8") as f:
    for line in f:
        print(line.strip())

Downloads last month: 57

Collection including kipasyangin5/5m-terminal-lm-dataset

5M Terminal Language Model Suite

Collection

Complete 5.0M parameter terminal LM suite: Chinchilla & saturated models, raw 50M token corpus, pure CLI dataset, and ChatGPT playground Space. • 5 items • Updated Aug 1