Text Generation
Transformers
Safetensors
Bulgarian
English
llama
conversational
text-generation-inference
Inference Endpoints
zolicsaki commited on
Commit
b868179
1 Parent(s): a9ade51

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +11 -7
README.md CHANGED
@@ -26,6 +26,7 @@ SambaLingo-Bulgarian-Chat is a human aligned chat model trained in Bulgarian and
26
  - **Language(s):** Bulgarian, English
27
  - **Finetuned from model:** [Llama 2](https://huggingface.co/meta-llama/Llama-2-7b-hf)
28
  - **Try this model:** [SambaLingo-chat-space](https://huggingface.co/spaces/sambanovasystems/SambaLingo-chat-space)
 
29
  - **Blog Post**: [sambalingo-open-source-language-experts](https://sambanova.ai/blog/sambalingo-open-source-language-experts)
30
 
31
  ## Getting Started
@@ -84,6 +85,9 @@ The DPO phase was done on the [ultrafeedback](https://huggingface.co/datasets/Hu
84
  ## Tokenizer Details
85
  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.
86
 
 
 
 
87
  ## Uses
88
  <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
89
 
@@ -126,12 +130,12 @@ We would like to give a special thanks to the following groups:
126
 
127
  ## Cite SambaLingo
128
  ```
129
- @software{sambalingo,
130
- title = {{SambaLingo: Open Source Language Experts}},
131
- author = {SambaNova Systems},
132
- url = {https://huggingface.co/sambanovasystems/SambaLingo-Bulgarian-Chat}
133
- month = {2},
134
- year = {2024},
135
- version = {1.0},
136
  }
137
  ```
 
26
  - **Language(s):** Bulgarian, English
27
  - **Finetuned from model:** [Llama 2](https://huggingface.co/meta-llama/Llama-2-7b-hf)
28
  - **Try this model:** [SambaLingo-chat-space](https://huggingface.co/spaces/sambanovasystems/SambaLingo-chat-space)
29
+ - **Paper:** [SambaLingo: Teaching Large Language Models New Languages](https://arxiv.org/abs/2404.05829)
30
  - **Blog Post**: [sambalingo-open-source-language-experts](https://sambanova.ai/blog/sambalingo-open-source-language-experts)
31
 
32
  ## Getting Started
 
85
  ## Tokenizer Details
86
  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.
87
 
88
+ ## Evaluation
89
+ For evaluation results see our paper: [SambaLingo: Teaching Large Language Models New Languages](https://arxiv.org/abs/2404.05829)
90
+
91
  ## Uses
92
  <!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
93
 
 
130
 
131
  ## Cite SambaLingo
132
  ```
133
+ @misc{csaki2024sambalingo,
134
+ title={SambaLingo: Teaching Large Language Models New Languages},
135
+ 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},
136
+ year={2024},
137
+ eprint={2404.05829},
138
+ archivePrefix={arXiv},
139
+ primaryClass={cs.CL}
140
  }
141
  ```