Sponsored by APIMart
Toolify Toolify logo
AI Models
open-llama-2-ko-7b

beomi / open-llama-2-ko-7b

Run on huggingface.co
huggingface.co
Total runs: 44
24-hour runs: -1
7-day runs: -15
30-day runs: -39
Model's Last Updated: December 27 2023
text-generation

Introduction of open-llama-2-ko-7b

Model Details of open-llama-2-ko-7b

Update Log

  • 2023.12.14: Initial Release of Open-Llama-2-Ko

Open-Llama-2-Ko 🦙🇰🇷

Open-Llama-2-Ko represents an advanced iteration of the Llama 2 model, featuring an expanded vocabulary and the inclusion of a Korean corpus for enhanced pretraining. Similar to its predecessor, Llama-2-Ko, this model operates within the range of generative text models, with parameter counts ranging from 7 billion to 70 billion. The focus of this repository is on the 7B pretrained version, designed to integrate seamlessly with the Hugging Face Transformers format.

The primary distinction between the Llama-2-Ko Series and Open-Llama-2-Ko lies in the dataset. Open-Llama-2-Ko exclusively utilizes publicly accessible Korean corpora, including sources such as AI Hub , Modu Corpus, 모두의 말뭉치 , and Korean Wikipedia .

As training was conducted solely with publicly available corpora, this model is open for unrestricted use by everyone, adhering to the MIT License*.

*MIT License under LLAMA 2 COMMUNITY LICENSE AGREEMENT

Model Details

Model Developers: Junbum Lee (Beomi)

Variations: Open-Llama-2-Ko will be available in different parameter sizes — 7B and 13B — along with various pretrained options.

Input: The model accepts only text input.

Output: The model produces text output exclusively.

Model Architecture:

Open-Llama-2-Ko is an auto-regressive language model that leverages an optimized transformer architecture derived from Llama-2.

Training Data Parameters Content Length GQA Tokens Learning Rate
Llama 2 A curated mix of Publicly Accessible Korean Corpora 7B 2k ✘ >15B* 5e -5

Training Corpus

The model was trained using selected datasets from AIHub and Modu Corpus. Detailed information about the training datasets is available below:

The final JSONL dataset used to train this model is approximately 61GB in size.

Total token count: Approximately 15 billion tokens (*using the expanded tokenizer. With the original Llama tokenizer, >60 billion tokens.)

Vocab Expansion

Model Name Vocabulary Size Description
Original Llama-2 32000 Sentencepiece BPE
Expanded Llama-2-Ko 46336 Sentencepiece BPE. Added Korean vocab and merges

Tokenizing "안녕하세요, 오늘은 날씨가 좋네요."

Model Tokens
Llama-2 ['▁', '안', '<0xEB>', '<0x85>', '<0x95>', '하', '세', '요', ',', '▁', '오', '<0xEB>', '<0x8A>', '<0x98>', '은', '▁', '<0xEB>', '<0x82>', '<0xA0>', '씨', '가', '▁', '<0xEC>', '<0xA2>', '<0x8B>', '<0xEB>', '<0x84>', '<0xA4>', '요']
Llama-2-Ko ['▁안녕', '하세요', ',', '▁오늘은', '▁날', '씨가', '▁좋네요']

Tokenizing "Llama 2: Open Foundation and Fine-Tuned Chat Models"

Model Tokens
Llama-2 ['▁L', 'l', 'ama', '▁', '2', ':', '▁Open', '▁Foundation', '▁and', '▁Fine', '-', 'T', 'un', 'ed', '▁Ch', 'at', '▁Mod', 'els']
Llama-2-Ko ['▁L', 'l', 'ama', '▁', '2', ':', '▁Open', '▁Foundation', '▁and', '▁Fine', '-', 'T', 'un', 'ed', '▁Ch', 'at', '▁Mod', 'els']

LICENSE

MIT License under LLAMA 2 COMMUNITY LICENSE AGREEMENT

Model Benchmark

LM Eval Harness - Korean (polyglot branch)

TBD

Citation

TBD

Acknowledgements

Runs of beomi open-llama-2-ko-7b on huggingface.co

44
Total runs
-1
24-hour runs
-1
3-day runs
-15
7-day runs
-39
30-day runs

More Information About open-llama-2-ko-7b huggingface.co Model

More open-llama-2-ko-7b license Visit here:

https://choosealicense.com/licenses/mit

open-llama-2-ko-7b huggingface.co

open-llama-2-ko-7b huggingface.co is an AI model on huggingface.co that provides open-llama-2-ko-7b's model effect (), which can be used instantly with this beomi open-llama-2-ko-7b model. huggingface.co supports a free trial of the open-llama-2-ko-7b model, and also provides paid use of the open-llama-2-ko-7b. Support call open-llama-2-ko-7b model through api, including Node.js, Python, http.

open-llama-2-ko-7b huggingface.co Url

https://huggingface.co/beomi/open-llama-2-ko-7b

beomi open-llama-2-ko-7b online free

open-llama-2-ko-7b huggingface.co is an online trial and call api platform, which integrates open-llama-2-ko-7b's modeling effects, including api services, and provides a free online trial of open-llama-2-ko-7b, you can try open-llama-2-ko-7b online for free by clicking the link below.

beomi open-llama-2-ko-7b online free url in huggingface.co:

https://huggingface.co/beomi/open-llama-2-ko-7b

open-llama-2-ko-7b install

open-llama-2-ko-7b is an open source model from GitHub that offers a free installation service, and any user can find open-llama-2-ko-7b on GitHub to install. At the same time, huggingface.co provides the effect of open-llama-2-ko-7b install, users can directly use open-llama-2-ko-7b installed effect in huggingface.co for debugging and trial. It also supports api for free installation.

open-llama-2-ko-7b install url in huggingface.co:

https://huggingface.co/beomi/open-llama-2-ko-7b

Url of open-llama-2-ko-7b

open-llama-2-ko-7b huggingface.co Url

Provider of open-llama-2-ko-7b huggingface.co

beomi
ORGANIZATIONS
huggingface.co

Other API from beomi

 huggingface.co
beomi/kcbert-base

Total runs: 4.1K
Run Growth: 524
Growth Rate: 12.94%
Updated:March 30 2023
huggingface.co
beomi/KcELECTRA-base

Total runs: 4.0K
Run Growth: -1.9K
Growth Rate: -48.61%
Updated:May 23 2024
huggingface.co
beomi/kcbert-large

Total runs: 686
Run Growth: 318
Growth Rate: 46.36%
Updated:March 30 2023
huggingface.co
beomi/Yi-Ko-6B

Total runs: 565
Run Growth: 19
Growth Rate: 3.36%
Updated:July 08 2024
huggingface.co
beomi/llama-2-ko-7b

Total runs: 525
Run Growth: -138
Growth Rate: -26.29%
Updated:December 27 2023
huggingface.co
beomi/kcgpt2

Total runs: 388
Run Growth: 353
Growth Rate: 90.98%
Updated:November 23 2021
huggingface.co
beomi/KoRWKV-6B

Total runs: 302
Run Growth: 122
Growth Rate: 40.40%
Updated:July 20 2023
huggingface.co
beomi/kollama-13b

Total runs: 267
Run Growth: 138
Growth Rate: 51.69%
Updated:June 28 2023
huggingface.co
beomi/Yi-Ko-DUS-9B

Total runs: 264
Run Growth: 18
Growth Rate: 6.82%
Updated:July 08 2024
huggingface.co
beomi/gemma-ko-7b

Total runs: 141
Run Growth: 13
Growth Rate: 9.22%
Updated:March 26 2024
huggingface.co
beomi/KoRWKV-1.5B

Total runs: 98
Run Growth: 49
Growth Rate: 50.00%
Updated:November 13 2023
huggingface.co
beomi/gemma-ko-2b

Total runs: 67
Run Growth: -129
Growth Rate: -192.54%
Updated:March 26 2024
huggingface.co
beomi/gemma-mling-7b

Total runs: 49
Run Growth: -6
Growth Rate: -12.24%
Updated:August 21 2024
huggingface.co
beomi/kobert

Total runs: 48
Run Growth: 19
Growth Rate: 39.58%
Updated:September 15 2023
huggingface.co
beomi/kollama-7b

Total runs: 32
Run Growth: -80
Growth Rate: -250.00%
Updated:July 11 2023
huggingface.co
beomi/Yi-Ko-34B

Total runs: 31
Run Growth: -5
Growth Rate: -16.13%
Updated:July 08 2024
huggingface.co
beomi/qwen3-8b-dmax

Total runs: 24
Run Growth: 18
Growth Rate: 78.26%
Updated:May 04 2026
huggingface.co
beomi/exKcBERT-paws

Total runs: 14
Run Growth: 4
Growth Rate: 28.57%
Updated:June 11 2021
huggingface.co
beomi/kcgpt2-dev

Total runs: 13
Run Growth: 1
Growth Rate: 7.69%
Updated:May 21 2021
huggingface.co
beomi/kollama-33b

Total runs: 11
Run Growth: 3
Growth Rate: 27.27%
Updated:June 29 2023
huggingface.co
beomi/KcT5-dev

Total runs: 9
Run Growth: -25
Growth Rate: -277.78%
Updated:March 01 2022
huggingface.co
beomi/KcRoBERTa-dev

Total runs: 8
Run Growth: 0
Growth Rate: 0.00%
Updated:March 08 2023
huggingface.co
beomi/KcBERT-v2023

Total runs: 7
Run Growth: -3
Growth Rate: -42.86%
Updated:May 07 2023
huggingface.co
beomi/llama-2-ko-70b

Total runs: 1
Run Growth: 1
Growth Rate: 100.00%
Updated:November 03 2023
huggingface.co
beomi/KcT5

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:February 10 2022
huggingface.co
beomi/exKcBERT

Total runs: 0
Run Growth: 0
Growth Rate: 0.00%
Updated:June 10 2021