Llama-2-Ko-Chat 🦙🇰🇷

<img src=https://github.com/boostcampaitech5/level2_klue-nlp-08/assets/96530685/9dc3a02a-f554-42c7-9aa7-e7bd8c8870b7 style="max-width: 700px; width: 100%" />

Llama-2-Ko-7b-Chat은 beomi/llama-2-ko-7b 40B를 토대로 만들어졌습니다. 학습 데이터는 nlpai-lab/kullm-v2를 통해 학습하였습니다. 아직 학습이 진행 중이며 추후 beomi/llama-2-ko-7b의 업데이트에 따라 추가로 훈련을 진행할 계획입니다. 이 모델은 Naver BoostCamp NLP-08 프로젝트를 토대로 만들어졌습니다.

프로젝트 링크 : LLaMA

Model Details

Backbone Model beomi/llama-2-ko-7b

Model Developers Taemin Kim (kfkas), Juwon Kim (uomnf97)

data nlpai-lab/kullm-v2

prompts kullm

Input Models input text only.

Output Models generate text only.

---

---

---

훈련 진행 현황

<img src=https://github.com/taemin6697/Paper_Review/assets/96530685/b9a697a2-ef06-4b1c-97e1-e72b20d9a8b5 style="max-width: 700px; width: 100%" />

Inference

def gen(x, model, tokenizer, device):
    prompt = (
        f"아래는 작업을 설명하는 명령어입니다. 요청을 적절히 완료하는 응답을 작성하세요.\n\n### 명령어:\n{x}\n\n### 응답:"
    )
    len_prompt = len(prompt)
    gened = model.generate(
        **tokenizer(prompt, return_tensors="pt", return_token_type_ids=False).to(
            device
        ),
        max_new_tokens=1024,
        early_stopping=True,
        do_sample=True,
        top_k=20,
        top_p=0.92,
        no_repeat_ngram_size=3,
        eos_token_id=2,
        repetition_penalty=1.2,
        num_beams=3
    )
    return tokenizer.decode(gened[0])[len_prompt:]

def LLM_infer(input):
    device = (
        torch.device("cuda:0") if torch.cuda.is_available() else torch.device("cpu")
    )
    model_id = "kfkas/Llama-2-ko-7b-Chat"
    model = AutoModelForCausalLM.from_pretrained(
        model_id, device_map={"": 0},torch_dtype=torch.float16, low_cpu_mem_usage=True
    )
    tokenizer = AutoTokenizer.from_pretrained(model_id)
    model.eval()
    model.config.use_cache = (True)
    tokenizer.pad_token = tokenizer.eos_token
    output = gen(input, model=model, tokenizer=tokenizer, device=device)

    return output


if __name__ == "__main__":
    text = LLM_infer("삼원색에 대해 알려줘")
    print(text)

Note for oobabooga/text-generation-webui

Remove ValueError at load_tokenizer function(line 109 or near), in modules/models.py.

diff --git a/modules/models.py b/modules/models.py
index 232d5fa..de5b7a0 100644
--- a/modules/models.py
+++ b/modules/models.py
@@ -106,7 +106,7 @@ def load_tokenizer(model_name, model):
                 trust_remote_code=shared.args.trust_remote_code,
                 use_fast=False
             )
-        except ValueError:
+        except:
             tokenizer = AutoTokenizer.from_pretrained(
                 path_to_model,
                 trust_remote_code=shared.args.trust_remote_code,

Since Llama-2-Ko uses FastTokenizer provided by HF tokenizers NOT sentencepiece package, it is required to use use_fast=True option when initialize tokenizer.

Apple Sillicon does not support BF16 computing, use CPU instead. (BF16 is supported when using NVIDIA GPU)

---

Below is the original model card of the Llama-2 model.

Llama 2

Llama 2 is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. This is the repository for the 7B pretrained model, converted for the Hugging Face Transformers format. Links to other models can be found in the index at the bottom.

Model Details

Note: Use of this model is governed by the Meta license. In order to download the model weights and tokenizer, please visit the website and accept our License before requesting access here.

Meta developed and publicly released the Llama 2 family of large language models (LLMs), a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama-2-Chat, are optimized for dialogue use cases. Llama-2-Chat models outperform open-source chat models on most benchmarks we tested, and in our human evaluations for helpfulness and safety, are on par with some popular closed-source models like ChatGPT and PaLM.

Model Developers Meta

Variations Llama 2 comes in a range of parameter sizes — 7B, 13B, and 70B — as well as pretrained and fine-tuned variations.

Input Models input text only.

Output Models generate text only.

Model Architecture Llama 2 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.

Llama 2 family of models. Token counts refer to pretraining data only. All models are trained with a global batch-size of 4M tokens. Bigger models - 70B -- use Grouped-Query Attention (GQA) for improved inference scalability.

Model Dates Llama 2 was trained between January 2023 and July 2023.

Status This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.

License A custom commercial license is available at: https://ai.meta.com/resources/models-and-libraries/llama-downloads/

Research Paper "Llama-2: Open Foundation and Fine-tuned Chat Models"

Intended Use

Intended Use Cases Llama 2 is intended for commercial and research use in English. Tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.

To get the expected features and performance for the chat versions, a specific formatting needs to be followed, including the INST and <<SYS>> tags, BOS and EOS tokens, and the whitespaces and breaklines in between (we recommend calling strip() on inputs to avoid double-spaces). See our reference code in github for details: chat_completion.

Out-of-scope Uses Use in any manner that violates applicable laws or regulations (including trade compliance laws).Use in languages other than English. Use in any other way that is prohibited by the Acceptable Use Policy and Licensing Agreement for Llama 2.

Hardware and Software

Training Factors We used custom training libraries, Meta's Research Super Cluster, and production clusters for pretraining. Fine-tuning, annotation, and evaluation were also performed on third-party cloud compute.

Carbon Footprint Pretraining utilized a cumulative 3.3M GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 539 tCO2eq, 100% of which were offset by Meta’s sustainability program.

CO₂ emissions during pretraining. Time: total GPU time required for training each model. Power Consumption: peak power capacity per GPU device for the GPUs used adjusted for power usage efficiency. 100% of the emissions are directly offset by Meta's sustainability program, and because we are openly releasing these models, the pretraining costs do not need to be incurred by others.

Training Data

Overview Llama 2 was pretrained on 2 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over one million new human-annotated examples. Neither the pretraining nor the fine-tuning datasets include Meta user data.

Data Freshness The pretraining data has a cutoff of September 2022, but some tuning data is more recent, up to July 2023.

Evaluation Results

In this section, we report the results for the Llama 1 and Llama 2 models on standard academic benchmarks.For all the evaluations, we use our internal evaluations library.

Overall performance on grouped academic benchmarks. Code: We report the average pass@1 scores of our models on HumanEval and MBPP. Commonsense Reasoning: We report the average of PIQA, SIQA, HellaSwag, WinoGrande, ARC easy and challenge, OpenBookQA, and CommonsenseQA. We report 7-shot results for CommonSenseQA and 0-shot results for all other benchmarks. World Knowledge: We evaluate the 5-shot performance on NaturalQuestions and TriviaQA and report the average. Reading Comprehension: For reading comprehension, we report the 0-shot average on SQuAD, QuAC, and BoolQ. MATH: We report the average of the GSM8K (8 shot) and MATH (4 shot) benchmarks at top 1.

Evaluation of pretrained LLMs on automatic safety benchmarks. For TruthfulQA, we present the percentage of generations that are both truthful and informative (the higher the better). For ToxiGen, we present the percentage of toxic generations (the smaller the better).

Evaluation of fine-tuned LLMs on different safety datasets. Same metric definitions as above.

Ethical Considerations and Limitations

Llama 2 is a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, Llama 2’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 2, developers should perform safety testing and tuning tailored to their specific applications of the model.

Please see the Responsible Use Guide available at https://ai.meta.com/llama/responsible-use-guide/

Reporting Issues

Please report any software “bug,” or other problems with the models through one of the following means:

• Reporting issues with the model: github.com/facebookresearch/llama
• Reporting problematic content generated by the model: developers.facebook.com/llama_output_feedback
• Reporting bugs and security concerns: facebook.com/whitehat/info

Llama Model Index