--- license: cc-by-nc-4.0 language: - ro base_model: - OpenLLM-Ro/RoLlama3.1-8b-Instruct-2024-10-09 datasets: - OpenLLM-Ro/ro_dpo_helpsteer model-index: - name: OpenLLM-Ro/RoLlama3.1-8b-Instruct-DPO-2024-10-09 results: - task: type: text-generation dataset: name: RoMT-Bench type: RoMT-Bench metrics: - name: Score type: Score value: 6.21 - task: type: text-generation dataset: name: RoCulturaBench type: RoCulturaBench metrics: - name: Score type: Score value: 4.42 - task: type: text-generation dataset: name: Romanian_Academic_Benchmarks type: Romanian_Academic_Benchmarks metrics: - name: Average accuracy type: accuracy value: 52.74 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_arc_challenge type: OpenLLM-Ro/ro_arc_challenge metrics: - name: Average accuracy type: accuracy value: 44.84 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_mmlu type: OpenLLM-Ro/ro_mmlu metrics: - name: Average accuracy type: accuracy value: 55.06 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_winogrande type: OpenLLM-Ro/ro_winogrande metrics: - name: Average accuracy type: accuracy value: 65.87 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_hellaswag type: OpenLLM-Ro/ro_hellaswag metrics: - name: Average accuracy type: accuracy value: 58.67 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_gsm8k type: OpenLLM-Ro/ro_gsm8k metrics: - name: Average accuracy type: accuracy value: 44.17 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_truthfulqa type: OpenLLM-Ro/ro_truthfulqa metrics: - name: Average accuracy type: accuracy value: 47.82 - task: type: text-generation dataset: name: LaRoSeDa_binary type: LaRoSeDa_binary metrics: - name: Average macro-f1 type: macro-f1 value: 96.10 - task: type: text-generation dataset: name: LaRoSeDa_multiclass type: LaRoSeDa_multiclass metrics: - name: Average macro-f1 type: macro-f1 value: 55.37 - task: type: text-generation dataset: name: LaRoSeDa_binary_finetuned type: LaRoSeDa_binary_finetuned metrics: - name: Average macro-f1 type: macro-f1 value: 0.00 - task: type: text-generation dataset: name: LaRoSeDa_multiclass_finetuned type: LaRoSeDa_multiclass_finetuned metrics: - name: Average macro-f1 type: macro-f1 value: 0.00 - task: type: text-generation dataset: name: WMT_EN-RO type: WMT_EN-RO metrics: - name: Average bleu type: bleu value: 21.29 - task: type: text-generation dataset: name: WMT_RO-EN type: WMT_RO-EN metrics: - name: Average bleu type: bleu value: 21.86 - task: type: text-generation dataset: name: WMT_EN-RO_finetuned type: WMT_EN-RO_finetuned metrics: - name: Average bleu type: bleu value: 0.00 - task: type: text-generation dataset: name: WMT_RO-EN_finetuned type: WMT_RO-EN_finetuned metrics: - name: Average bleu type: bleu value: 0.00 - task: type: text-generation dataset: name: XQuAD type: XQuAD metrics: - name: Average exact_match type: exact_match value: 21.58 - task: type: text-generation dataset: name: XQuAD type: XQuAD metrics: - name: Average f1 type: f1 value: 36.54 - task: type: text-generation dataset: name: XQuAD_finetuned type: XQuAD_finetuned metrics: - name: Average exact_match type: exact_match value: 0.00 - task: type: text-generation dataset: name: XQuAD_finetuned type: XQuAD_finetuned metrics: - name: Average f1 type: f1 value: 0.00 - task: type: text-generation dataset: name: STS type: STS metrics: - name: Average spearman type: spearman value: 78.01 - task: type: text-generation dataset: name: STS type: STS metrics: - name: Average pearson type: pearson value: 77.98 - task: type: text-generation dataset: name: STS_finetuned type: STS_finetuned metrics: - name: Average spearman type: spearman value: 0.00 - task: type: text-generation dataset: name: STS_finetuned type: STS_finetuned metrics: - name: Average pearson type: pearson value: 0.00 - task: type: text-generation dataset: name: RoMT-Bench type: RoMT-Bench metrics: - name: First turn type: Score value: 6.74 - name: Second turn type: Score value: 5.69 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_arc_challenge type: OpenLLM-Ro/ro_arc_challenge metrics: - name: 0-shot type: accuracy value: 41.82 - name: 1-shot type: accuracy value: 43.70 - name: 3-shot type: accuracy value: 45.33 - name: 5-shot type: accuracy value: 46.10 - name: 10-shot type: accuracy value: 45.76 - name: 25-shot type: accuracy value: 46.36 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_mmlu type: OpenLLM-Ro/ro_mmlu metrics: - name: 0-shot type: accuracy value: 53.75 - name: 1-shot type: accuracy value: 54.94 - name: 3-shot type: accuracy value: 56.07 - name: 5-shot type: accuracy value: 55.47 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_winogrande type: OpenLLM-Ro/ro_winogrande metrics: - name: 0-shot type: accuracy value: 64.40 - name: 1-shot type: accuracy value: 66.22 - name: 3-shot type: accuracy value: 65.75 - name: 5-shot type: accuracy value: 67.09 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_hellaswag type: OpenLLM-Ro/ro_hellaswag metrics: - name: 0-shot type: accuracy value: 57.25 - name: 1-shot type: accuracy value: 58.00 - name: 3-shot type: accuracy value: 59.23 - name: 5-shot type: accuracy value: 59.30 - name: 10-shot type: accuracy value: 59.56 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_gsm8k type: OpenLLM-Ro/ro_gsm8k metrics: - name: 1-shot type: accuracy value: 36.47 - name: 3-shot type: accuracy value: 45.94 - name: 5-shot type: accuracy value: 50.11 - task: type: text-generation dataset: name: LaRoSeDa_binary type: LaRoSeDa_binary metrics: - name: 0-shot type: macro-f1 value: 93.11 - name: 1-shot type: macro-f1 value: 96.06 - name: 3-shot type: macro-f1 value: 97.53 - name: 5-shot type: macro-f1 value: 97.70 - task: type: text-generation dataset: name: LaRoSeDa_multiclass type: LaRoSeDa_multiclass metrics: - name: 0-shot type: macro-f1 value: 65.61 - name: 1-shot type: macro-f1 value: 55.73 - name: 3-shot type: macro-f1 value: 46.33 - name: 5-shot type: macro-f1 value: 53.82 - task: type: text-generation dataset: name: WMT_EN-RO type: WMT_EN-RO metrics: - name: 0-shot type: bleu value: 6.89 - name: 1-shot type: bleu value: 26.62 - name: 3-shot type: bleu value: 25.70 - name: 5-shot type: bleu value: 25.94 - task: type: text-generation dataset: name: WMT_RO-EN type: WMT_RO-EN metrics: - name: 0-shot type: bleu value: 2.16 - name: 1-shot type: bleu value: 16.65 - name: 3-shot type: bleu value: 33.41 - name: 5-shot type: bleu value: 35.22 - task: type: text-generation dataset: name: XQuAD_EM type: XQuAD_EM metrics: - name: 0-shot type: exact_match value: 8.99 - name: 1-shot type: exact_match value: 35.88 - name: 3-shot type: exact_match value: 31.26 - name: 5-shot type: exact_match value: 10.17 - task: type: text-generation dataset: name: XQuAD_F1 type: XQuAD_F1 metrics: - name: 0-shot type: f1 value: 20.00 - name: 1-shot type: f1 value: 59.41 - name: 3-shot type: f1 value: 48.41 - name: 5-shot type: f1 value: 18.33 - task: type: text-generation dataset: name: STS_Spearman type: STS_Spearman metrics: - name: 1-shot type: spearman value: 78.10 - name: 3-shot type: spearman value: 77.81 - name: 5-shot type: spearman value: 78.11 - task: type: text-generation dataset: name: STS_Pearson type: STS_Pearson metrics: - name: 1-shot type: pearson value: 78.30 - name: 3-shot type: pearson value: 77.58 - name: 5-shot type: pearson value: 78.06 --- # Model Card for Model ID *Built with Meta Llama 3.1* This model points/is identical to [RoLlama3.1-8b-Instruct-DPO-2024-10-09](https://huggingface.co/OpenLLM-Ro/RoLlama3.1-8b-Instruct-DPO-2024-10-09). RoLlama3.1 is a family of pretrained and fine-tuned generative text models for Romanian. This is the repository for the **human aligned instruct 8B model**. Links to other models can be found at the bottom of this page. ## Model Details ### Model Description OpenLLM-Ro represents the first open-source effort to build a LLM specialized for Romanian. OpenLLM-Ro developed and publicly releases a collection of Romanian LLMs, both in the form of foundational model and instruct and chat variants. - **Developed by:** OpenLLM-Ro - **Language(s):** Romanian - **License:** cc-by-nc-4.0 - **Finetuned from model:** [RoLlama3.1-8b-Instruct-2024-10-09](https://huggingface.co/OpenLLM-Ro/RoLlama3.1-8b-Instruct-2024-10-09) - **Trained using:** [RoHelpSteer](https://huggingface.co/datasets/OpenLLM-Ro/ro_dpo_helpsteer) ### Model Sources - **Repository:** https://github.com/OpenLLM-Ro/LLaMA-Factory - **Paper:** https://arxiv.org/abs/2406.18266 ## Intended Use ### Intended Use Cases RoLlama3.1 is intented for research use in Romanian. Base models can be adapted for a variety of natural language tasks while instruction and chat tuned models are intended for assistant-like chat. ### Out-of-Scope Use Use in any manner that violates the license, any applicable laws or regluations, use in languages other than Romanian. ## How to Get Started with the Model Use the code below to get started with the model. ```python from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("OpenLLM-Ro/RoLlama3.1-8b-Instruct-DPO") model = AutoModelForCausalLM.from_pretrained("OpenLLM-Ro/RoLlama3.1-8b-Instruct-DPO") instruction = "Ce jocuri de societate pot juca cu prietenii mei?" chat = [ {"role": "system", "content": "Ești un asistent folositor, respectuos și onest. Încearcă să ajuți cât mai mult prin informațiile oferite, excluzând răspunsuri toxice, rasiste, sexiste, periculoase și ilegale."}, {"role": "user", "content": instruction}, ] prompt = tokenizer.apply_chat_template(chat, tokenize=False, system_message="") inputs = tokenizer.encode(prompt, add_special_tokens=False, return_tensors="pt") outputs = model.generate(input_ids=inputs, max_new_tokens=128) print(tokenizer.decode(outputs[0])) ``` ## Academic Benchmarks
Model
Average
ARC
MMLU
Winogrande
Hellaswag
GSM8k
TruthfulQA
Llama-3.1-8B-Instruct
49.87
42.86
53.73
59.71
56.82
35.56
50.54
RoLlama3.1-8b-Instruct-2024-10-09
53.03
47.69
54.57
65.84
59.94
44.30
45.82
RoLlama3.1-8b-Instruct-DPO-2024-10-09
52.74
44.84
55.06
65.87
58.67
44.17
47.82
## Downstream tasks
LaRoSeDa
WMT
Few-shot
Finetuned
Few-shot
Finetuned
Model
Binary
(Macro F1)
Multiclass
(Macro F1)
Binary
(Macro F1)
Multiclass
(Macro F1)
EN-RO
(Bleu)
RO-EN
(Bleu)
EN-RO
(Bleu)
RO-EN
(Bleu)
Llama-3.1-8B-Instruct
95.74
59.49
98.57
82.41
19.01
27.77
29.02
39.80
RoLlama3.1-8b-Instruct-2024-10-09
94.56
60.10
95.12
87.53
21.88
23.99
28.27
40.44
RoLlama3.1-8b-Instruct-DPO-2024-10-09
96.10
55.37
-
-
21.29
21.86
-
-
XQuAD
STS
Few-shot
Finetuned
Few-shot
Finetuned
Model
(EM)
(F1)
(EM)
(F1)
(Spearman)
(Pearson)
(Spearman)
(Pearson)
Llama-3.1-8B-Instruct
44.96
64.45
69.50
84.31
72.11
71.64
84.59
84.96
RoLlama3.1-8b-Instruct-2024-10-09
13.59
23.56
49.41
62.93
75.89
76.00
86.86
87.05
RoLlama3.1-8b-Instruct-DPO-2024-10-09
21.58
36.54
-
-
78.01
77.98
-
-
## MT-Bench
Model
Average
1st turn
2nd turn
Answers in Ro
Llama-3.1-8B-Instruct
5.69
5.85
5.53
160/160
RoLlama3.1-8b-Instruct-2024-10-09
5.42
5.95
4.89
160/160
RoLlama3.1-8b-Instruct-DPO-2024-10-09
6.21
6.74
5.69
160/160
## RoCulturaBench
Model
Average
Answers in Ro
Llama-3.1-8B-Instruct
3.54
100/100
RoLlama3.1-8b-Instruct-2024-10-09
3.55
100/100
RoLlama3.1-8b-Instruct-DPO-2024-10-09
4.42
100/100
## RoLlama3.1 Model Family | Model | Link | |--------------------|:--------:| |RoLlama3.1-8b-Instruct-2024-10-09| [link](https://huggingface.co/OpenLLM-Ro/RoLlama3.1-8b-Instruct-2024-10-09) | |*RoLlama3.1-8b-Instruct-DPO-2024-10-09*| [link](https://huggingface.co/OpenLLM-Ro/RoLlama3.1-8b-Instruct-DPO-2024-10-09) | ## Citation ``` @misc{masala2024vorbecstiromanecsterecipetrain, title={"Vorbe\c{s}ti Rom\^ane\c{s}te?" A Recipe to Train Powerful Romanian LLMs with English Instructions}, author={Mihai Masala and Denis C. Ilie-Ablachim and Alexandru Dima and Dragos Corlatescu and Miruna Zavelca and Ovio Olaru and Simina Terian-Dan and Andrei Terian-Dan and Marius Leordeanu and Horia Velicu and Marius Popescu and Mihai Dascalu and Traian Rebedea}, year={2024}, eprint={2406.18266}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2406.18266}, } ```