RichardErkhov/sambanovasystems_-_SambaLingo-Russian-Chat-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

SambaLingo-Russian-Chat - GGUF

• —Model creator: https://huggingface.co/sambanovasystems/
• —Original model: https://huggingface.co/sambanovasystems/SambaLingo-Russian-Chat/

Original model description: --- license: llama2 datasets:

• —HuggingFaceH4/ultrachat_200k
• —HuggingFaceH4/ultrafeedback_binarized
• —HuggingFaceH4/cai-conversation-harmless language:
• —ru
• —en ---

SambaLingo-Russian-Chat

<img src="SambaLingo_Logo.png" width="340" style="margin-left:'auto' margin-right:'auto' display:'block'"/>

<!-- Provide a quick summary of what the model is/does. --> SambaLingo-Russian-Chat is a human aligned chat model trained in Russian and English. It is trained using direct preference optimization on top the base model SambaLingo-Russian-Base. The base model adapts Llama-2-7b to Russian by training on 63 billion tokens from the Russian split of the Cultura-X dataset. Try this model at SambaLingo-chat-space.

Model Description

<!-- Provide a longer summary of what this model is. -->

• —Developed by: SambaNova Systems
• —Model type: Language Model
• —Language(s): Russian, English
• —Finetuned from model: Llama-2-7b
• —Try this model: SambaLingo-chat-space
• —Paper: SambaLingo: Teaching Large Language Models New Languages
• —Blog Post: sambalingo-open-source-language-experts

Getting Started

Loading Model With Hugging Face

Please make sure to set use_fast=False when loading the tokenizer.

from transformers import AutoModelForCausalLM, AutoTokenizer

tokenizer = AutoTokenizer.from_pretrained("sambanovasystems/SambaLingo-Russian-Chat", use_fast=False)
model = AutoModelForCausalLM.from_pretrained("sambanovasystems/SambaLingo-Russian-Chat", device_map="auto", torch_dtype="auto")

Interacting With Model Pipeline

Please make sure to set use_fast=False when loading the tokenizer.

from transformers import pipeline
pipe = pipeline("text-generation", model="sambanovasystems/SambaLingo-Russian-Chat", device_map="auto", use_fast=False)
messages = [
                {"role": "user", "content": {YOUR_QUESTION}},
]
prompt = pipe.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipe(prompt)[0]
outputs = outputs["generated_text"]

Suggested Inference Parameters

• —Temperature: 0.8
• —Repetition penalty: 1.0
• —Top-p: 0.9

Prompting Guidelines

To prompt this model, please use the following chat template:

<|user|>\n{question}</s>\n<|assistant|>\n

Example Prompts and Generations

<|user|>
Завершите параграф: Родион Раскольников - главный персонаж романа "Преступление и наказание" Федора Достоевского. Его характер очень сложен и     эволюционирует на протяжении всей истории. Он обладает выдающимися умственными способностями и амбициями, но</s>
<|assistant|>
его внутренняя бор    ьба и моральные сомнения мешают ему действовать в соответствии со своими желаниями. Раскольников — бедный студент, который терпит финансовые трудности и чувствует себ    я бессильным изменить свою жизнь. Он приходит к теории, согласно которой некоторые люди имеют право совершать преступления ради достижения великих целей, и решает про    верить свою теорию, убив старуху-процентщицу. Однако его действия приводят к глубокому чувству вины и психологическому срыву, который заставляет его искать искупления     и искупления. На протяжении всего романа Раскольников проходит путь самопознания и личностного роста, осознавая важность сочувствия, сострадания и ответственности пе    ред другими. В конечном итоге он принимает на себя ответственность за свои действия и находит искупление через раскаяние и прощение.

Training Details

The alignment phase follows the recipe for Zephyr-7B, and comprises two stages: supervised fine-tuning (SFT) and Direct Performance Optimization (DPO).

The SFT phase was done on the ultrachat_200k dataset mixed with the Google translated version of the ultrachat_200k dataset. It was trained for one epoch with global batch size 512 and max sequence length 2048 tokens. We used a linear decay learning rate of 2e-5 and 10% warmup.

The DPO phase was done on the ultrafeedback dataset and cai-conversation-harmless dataset, mixed with 10% of the data Google translated. It was trained with global batch size 32 and for three epochs. We used a linear decay learning rate of 5e-7, 10% warmup and β=0.1 as the regularization factor for DPO.

Tokenizer Details

We extended the vocabulary of the base llama model from 32,000 tokens to 57,000 tokens by adding up to 25,000 non-overlapping tokens from the new language.

Evaluation

For evaluation results see our paper: SambaLingo: Teaching Large Language Models New Languages

Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

Direct Use

<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. --> Use of this model is governed by the Meta’s Llama 2 Community License Agreement. Please review and accept the license before downloading the model weights.

Out-of-Scope Use

<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. --> SambaLingo should NOT be used for:

• —Mission-critical applications
• —Applications that involve the safety of others
• —Making highly important decisions

Bias, Risks, and Limitations

<!-- This section is meant to convey both technical and sociotechnical limitations. -->

Like all LLMs, SambaLingo has certain limitations:

• —Hallucination: Model may sometimes generate responses that contain plausible-sounding but factually incorrect or irrelevant information.
• —Code Switching: The model might unintentionally switch between languages or dialects within a single response, affecting the coherence and understandability of the output.
• —Repetition: The Model may produce repetitive phrases or sentences, leading to less engaging and informative responses.
• —Coding and Math: The model's performance in generating accurate code or solving complex mathematical problems may be limited.
• —Toxicity: The model could inadvertently generate responses containing inappropriate or harmful content.

Acknowledgments

We extend our heartfelt gratitude to the open-source AI community; this endeavor would not have been possible without open source. SambaNova embraces the open-source community and aspires to actively contribute to this initiative.

We would like to give a special thanks to the following groups:

• —Meta for open sourcing LLama 2 and open sourcing FLORES-200 dataset
• —Nguyen et al for open sourcing CulturaX dataset
• —CohereAI for releasing AYA-101 and open sourcing a multilingual instruction tuning dataset
• —EleutherAI for their open source evaluation framework
• —Hugging Face-H4 team for open source the zephyr training recipe and alignment handbook repo

Cite SambaLingo

@misc{csaki2024sambalingo,
      title={SambaLingo: Teaching Large Language Models New Languages}, 
      author={Zoltan Csaki and Bo Li and Jonathan Li and Qiantong Xu and Pian Pawakapan and Leon Zhang and Yun Du and Hengyu Zhao and Changran Hu and Urmish Thakker},
      year={2024},
      eprint={2404.05829},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}