word
stringlengths
5
33
target
stringclasses
81 values
count
int64
1
4
มิหนำซ้ำ
1
กระหมุดกระหมิด
2
เซอร์รารี่
3
จุตูปปาตญาณ
1
ลาผนวช
1
ยุทธวินัย
1
หมัดสุนัข
1
แก่วัด
1
นินนาท
1
กล้วยบวชชี
2
จิตใต้สำนีก
1
ปูชนียสถาน
1
กระแอบ
1
ง่าเงย
2
เมษายน
1
ผู้รู้แจ้ง
2
จีบพลู
1
พืชมงคล
1
สมเด็จพระญาณสังวร
2
กรรมวาจก
1
สนามไฟฟ้า
1
รับได้
1
แตกเนื้อหนุ่ม
1
ความรำคาญ
1
อาคารบ้านเรือน
3
เต้นท์
1
ขบวนแห่
1
ขอคืนดี
1
สิ่งยั่วยวน
1
ไพร่สม
1
สเปกโทรสโคป
1
ล็อบบี้
1
ผู้เข้ารับการอบรม
1
อุทกวิทยา
1
ทิโอฟิลีน
1
ไผ่เผ็ด
1
สั่งเสีย
2
ภูเบนทร์
1
สูดลูกหูลูกตา
4
พอๆกัน
1
ราศีกรกฎ
2
กัวรานา
1
เจ้านาย
1
ดื้อไม้
2
เกเรเกเส
1
ใจแป้ว
1
ข้อใหญ่ใจความ
1
ญาติห่างๆ
1
มาธุสร
1
ทะเบียนสำมะโนครัว
1
ทางไกล
1
จะละหวั่น
1
ซากพืช
1
อภิญญาณ
2
เมกะเฮิรตซ์
1
ฟูมฟัก
2
คาร์บอนไดออกไซด์
3
ลองจิจูด
2
คอนเฟิร์ม
1
ที่ต่ำที่สูง
3
แอสพาร์แตม
1
มีเงินทอง
1
พ.อ.พิเศษ
2
รถวิทยุ
1
สำนักข่าว
1
วิทยาศาสตร์กายภาพ
2
เจ้าเล่ห์แสนกล
1
กลับลำ
1
ตะลึงพรึงเพริด
2
ซาวนด์
1
จังหัน
1
อสุเรศ
1
แสร้งว่า
1
ตะปูจีน
1
แมลงแคง
2
ธงชัยราชกระบี่ยุทธ
1
หิตานุหิตประโยชน์
1
เบี้ยโบก
1
ซ่อมแซม
2
ชั่วพริบตาเดียว
2
พรรคประชาชนปฏิรูป
3
ใช้การไม่ได้
2
พิเชษฐ์
1
หยุนหนัน
1
ธารณามัย
1
จิตหลอน
1
แฮคเกอร์
1
เป็นประกาย
1
ขนมจีนซาวน้ำ
1
ก้านต่อดอก
2
ข่าวสั้น
1
มองไม่เห็น
2
ศิริพงษ์
1
นิพัทธ์
1
การลุก
2
ทีหน้า
1
ฮึกโหม
1
ผันผาย
1
อุกฉกรรจ์
2
สุไหงบูโล
1

Token Awareness Dataset

Built from "strawberry" meme 🍓, most LLM can't count the character of the word. Since this dataset is very easy to generate, we create a dataset for this benchmark just for fun.

For this dataset, we support only Thai (th) and English (en) language.

Dataset Creation

We sample words for each language from these sources:

We then sample 500 word each weigted by the word score, which was computed using this simple heuristic: S(w)=w1wi=1w1freq(ui)+j=1w2freq(bj) S(w) = \frac{\sqrt{|w|}}{\frac{1}{|w|} \sum_{i=1}^{|w|-1} \text{freq}(u_i) + \sum_{j=1}^{|w|-2} \text{freq}(b_j)}

where:

  • $|w|$ is the length of the word.
  • $u_i$ represents the unigram (single character) in the word.
  • $b_j$ represents the bigram (pair of consecutive characters) in the word.
  • $\text{freq}(u_i)$ is the frequency of unigram $u_i$ in the overall corpus.
  • $\text{freq}(b_j)$ is the frequency of bigram $b_j$ in the overall corpus.

we use a random seed of 42.

Author

Chompakorn Chaksangchaichot

Downloads last month
2
Edit dataset card