--- license: cc-by-nc-4.0 language: - ro base_model: - google/gemma-7b datasets: - OpenLLM-Ro/ro_sft_alpaca - OpenLLM-Ro/ro_sft_alpaca_gpt4 - OpenLLM-Ro/ro_sft_dolly - OpenLLM-Ro/ro_sft_selfinstruct_gpt4 - OpenLLM-Ro/ro_sft_norobots - OpenLLM-Ro/ro_sft_orca - OpenLLM-Ro/ro_sft_camel model-index: - name: OpenLLM-Ro/RoGemma-7b-Instruct-2024-06-28 results: - task: type: text-generation dataset: name: RoMT-Bench type: RoMT-Bench metrics: - name: Score type: Score value: 5.26 - task: type: text-generation dataset: name: RoCulturaBench type: RoCulturaBench metrics: - name: Score type: Score value: 3.26 - task: type: text-generation dataset: name: Romanian_Academic_Benchmarks type: Romanian_Academic_Benchmarks metrics: - name: Average accuracy type: accuracy value: 53.41 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_arc_challenge type: OpenLLM-Ro/ro_arc_challenge metrics: - name: Average accuracy type: accuracy value: 52.44 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_mmlu type: OpenLLM-Ro/ro_mmlu metrics: - name: Average accuracy type: accuracy value: 54.44 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_winogrande type: OpenLLM-Ro/ro_winogrande metrics: - name: Average accuracy type: accuracy value: 69.36 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_hellaswag type: OpenLLM-Ro/ro_hellaswag metrics: - name: Average accuracy type: accuracy value: 61.96 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_gsm8k type: OpenLLM-Ro/ro_gsm8k metrics: - name: Average accuracy type: accuracy value: 31.06 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_truthfulqa type: OpenLLM-Ro/ro_truthfulqa metrics: - name: Average accuracy type: accuracy value: 51.23 - task: type: text-generation dataset: name: LaRoSeDa_binary type: LaRoSeDa_binary metrics: - name: Average macro-f1 type: macro-f1 value: 97.86 - task: type: text-generation dataset: name: LaRoSeDa_multiclass type: LaRoSeDa_multiclass metrics: - name: Average macro-f1 type: macro-f1 value: 65.70 - task: type: text-generation dataset: name: LaRoSeDa_binary_finetuned type: LaRoSeDa_binary_finetuned metrics: - name: Average macro-f1 type: macro-f1 value: 98.43 - task: type: text-generation dataset: name: LaRoSeDa_multiclass_finetuned type: LaRoSeDa_multiclass_finetuned metrics: - name: Average macro-f1 type: macro-f1 value: 87.17 - task: type: text-generation dataset: name: WMT_EN-RO type: WMT_EN-RO metrics: - name: Average bleu type: bleu value: 27.91 - task: type: text-generation dataset: name: WMT_RO-EN type: WMT_RO-EN metrics: - name: Average bleu type: bleu value: 23.08 - task: type: text-generation dataset: name: WMT_EN-RO_finetuned type: WMT_EN-RO_finetuned metrics: - name: Average bleu type: bleu value: 27.99 - task: type: text-generation dataset: name: WMT_RO-EN_finetuned type: WMT_RO-EN_finetuned metrics: - name: Average bleu type: bleu value: 39.51 - task: type: text-generation dataset: name: XQuAD type: XQuAD metrics: - name: Average exact_match type: exact_match value: 17.75 - task: type: text-generation dataset: name: XQuAD type: XQuAD metrics: - name: Average f1 type: f1 value: 28.11 - task: type: text-generation dataset: name: XQuAD_finetuned type: XQuAD_finetuned metrics: - name: Average exact_match type: exact_match value: 52.02 - task: type: text-generation dataset: name: XQuAD_finetuned type: XQuAD_finetuned metrics: - name: Average f1 type: f1 value: 68.43 - task: type: text-generation dataset: name: STS type: STS metrics: - name: Average spearman type: spearman value: 73.96 - task: type: text-generation dataset: name: STS type: STS metrics: - name: Average pearson type: pearson value: 75.16 - task: type: text-generation dataset: name: STS_finetuned type: STS_finetuned metrics: - name: Average spearman type: spearman value: 86.45 - task: type: text-generation dataset: name: STS_finetuned type: STS_finetuned metrics: - name: Average pearson type: pearson value: 86.31 - task: type: text-generation dataset: name: RoMT-Bench type: RoMT-Bench metrics: - name: First turn type: Score value: 5.92 - name: Second turn type: Score value: 4.60 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_arc_challenge type: OpenLLM-Ro/ro_arc_challenge metrics: - name: 0-shot type: accuracy value: 50.30 - name: 1-shot type: accuracy value: 50.90 - name: 3-shot type: accuracy value: 52.53 - name: 5-shot type: accuracy value: 53.30 - name: 10-shot type: accuracy value: 54.33 - name: 25-shot type: accuracy value: 53.30 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_mmlu type: OpenLLM-Ro/ro_mmlu metrics: - name: 0-shot type: accuracy value: 54.95 - name: 1-shot type: accuracy value: 54.01 - name: 3-shot type: accuracy value: 54.03 - name: 5-shot type: accuracy value: 54.76 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_winogrande type: OpenLLM-Ro/ro_winogrande metrics: - name: 0-shot type: accuracy value: 68.67 - name: 1-shot type: accuracy value: 69.46 - name: 3-shot type: accuracy value: 68.43 - name: 5-shot type: accuracy value: 70.88 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_hellaswag type: OpenLLM-Ro/ro_hellaswag metrics: - name: 0-shot type: accuracy value: 61.54 - name: 1-shot type: accuracy value: 61.54 - name: 3-shot type: accuracy value: 62.08 - name: 5-shot type: accuracy value: 62.12 - name: 10-shot type: accuracy value: 62.51 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_gsm8k type: OpenLLM-Ro/ro_gsm8k metrics: - name: 1-shot type: accuracy value: 24.79 - name: 3-shot type: accuracy value: 34.50 - name: 5-shot type: accuracy value: 33.89 - task: type: text-generation dataset: name: LaRoSeDa_binary type: LaRoSeDa_binary metrics: - name: 0-shot type: macro-f1 value: 97.60 - name: 1-shot type: macro-f1 value: 97.23 - name: 3-shot type: macro-f1 value: 98.13 - name: 5-shot type: macro-f1 value: 98.50 - task: type: text-generation dataset: name: LaRoSeDa_multiclass type: LaRoSeDa_multiclass metrics: - name: 0-shot type: macro-f1 value: 68.53 - name: 1-shot type: macro-f1 value: 64.84 - name: 3-shot type: macro-f1 value: 63.62 - name: 5-shot type: macro-f1 value: 65.83 - task: type: text-generation dataset: name: WMT_EN-RO type: WMT_EN-RO metrics: - name: 0-shot type: bleu value: 25.04 - name: 1-shot type: bleu value: 28.43 - name: 3-shot type: bleu value: 28.87 - name: 5-shot type: bleu value: 29.28 - task: type: text-generation dataset: name: WMT_RO-EN type: WMT_RO-EN metrics: - name: 0-shot type: bleu value: 4.94 - name: 1-shot type: bleu value: 25.33 - name: 3-shot type: bleu value: 30.87 - name: 5-shot type: bleu value: 31.19 - task: type: text-generation dataset: name: XQuAD_EM type: XQuAD_EM metrics: - name: 0-shot type: exact_match value: 36.47 - name: 1-shot type: exact_match value: 26.22 - name: 3-shot type: exact_match value: 3.19 - name: 5-shot type: exact_match value: 5.13 - task: type: text-generation dataset: name: XQuAD_F1 type: XQuAD_F1 metrics: - name: 0-shot type: f1 value: 56.83 - name: 1-shot type: f1 value: 38.53 - name: 3-shot type: f1 value: 6.88 - name: 5-shot type: f1 value: 10.19 - task: type: text-generation dataset: name: STS_Spearman type: STS_Spearman metrics: - name: 1-shot type: spearman value: 70.61 - name: 3-shot type: spearman value: 73.53 - name: 5-shot type: spearman value: 77.73 - task: type: text-generation dataset: name: STS_Pearson type: STS_Pearson metrics: - name: 1-shot type: pearson value: 72.28 - name: 3-shot type: pearson value: 74.46 - name: 5-shot type: pearson value: 78.75 --- # Model Card for Model ID RoGemma is a family of pretrained and fine-tuned generative text models for Romanian. This is the repository for the **instruct 7B 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:** [gemma-7b](https://huggingface.co/google/gemma-7b) - **Trained using:** [RoAlpaca](https://huggingface.co/datasets/OpenLLM-Ro/ro_sft_alpaca), [RoAlpacaGPT4](https://huggingface.co/datasets/OpenLLM-Ro/ro_sft_alpaca_gpt4), [RoDolly](https://huggingface.co/datasets/OpenLLM-Ro/ro_sft_dolly), [RoSelfInstruct](https://huggingface.co/datasets/OpenLLM-Ro/ro_sft_selfinstruct_gpt4), [RoNoRobots](https://huggingface.co/datasets/OpenLLM-Ro/ro_sft_norobots), [RoOrca](https://huggingface.co/datasets/OpenLLM-Ro/ro_sft_orca), [RoCamel](https://huggingface.co/datasets/OpenLLM-Ro/ro_sft_camel). ### Model Sources - **Repository:** https://github.com/OpenLLM-Ro/LLaMA-Factory - **Paper:** https://arxiv.org/abs/2406.18266 ## Intended Use ### Intended Use Cases RoGemma 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/RoGemma-7b-Instruct-2024-06-28") model = AutoModelForCausalLM.from_pretrained("OpenLLM-Ro/RoGemma-7b-Instruct-2024-06-28") instruction = "Ce jocuri de societate pot juca cu prietenii mei?" chat = [ {"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
gemma-1.1-7b-it
41.44
40.32
47.22
55.01
47.03
9.50
49.58
RoGemma-7b-Instruct-2024-06-28
53.41
52.44
54.44
69.36
61.96
31.06
51.23
RoGemma-7b-Instruct-2024-10-09
50.48
52.01
52.37
66.97
56.34
25.98
49.18
RoGemma-7b-Instruct-DPO-2024-10-09
48.27
46.66
54.45
63.73
49.33
34.98
40.45
## 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)
gemma-1.1-7b-it
87.54
51.48
83.87
85.61
17.96
27.74
25.48
36.11
RoGemma-7b-Instruct-2024-06-28
97.86
65.70
98.43
87.17
27.91
23.08
27.99
39.51
RoGemma-7b-Instruct-2024-10-09
86.96
56.72
98.80
85.81
24.45
14.20
25.96
39.07
RoGemma-7b-Instruct-DPO-2024-10-09
96.45
63.23
-
-
20.73
7.87
-
-
XQuAD
STS
Few-shot
Finetuned
Few-shot
Finetuned
Model
(EM)
(F1)
(EM)
(F1)
(Spearman)
(Pearson)
(Spearman)
(Pearson)
gemma-1.1-7b-it
42.10
62.30
60.34
77.40
49.10
50.23
83.43
83.64
RoGemma-7b-Instruct-2024-06-28
17.75
28.11
52.02
68.43
73.96
75.16
86.45
86.31
RoGemma-7b-Instruct-2024-10-09
26.03
41.58
46.72
60.79
73.23
71.58
88.42
88.45
RoGemma-7b-Instruct-DPO-2024-10-09
19.14
38.10
-
-
69.38
69.34
-
-
## MT-Bench
Model
Average
1st turn
2nd turn
Answers in Ro
gemma-1.1-7b-it
4.83
5.11
4.55
160/160
RoGemma-7b-Instruct-2024-06-28
5.26
5.92
4.60
160/160
RoGemma-7b-Instruct-2024-10-09
5.24
5.55
4.94
160/160
RoGemma-7b-Instruct-DPO-2024-10-09
5.47
5.92
5.03
160/160
## RoCulturaBench
Model
Average
Answers in Ro
gemma-1.1-7b-it
3.38
100/100
RoGemma-7b-Instruct-2024-06-28
3.26
100/100
RoGemma-7b-Instruct-2024-10-09
3.51
100/100
RoGemma-7b-Instruct-DPO-2024-10-09
3.94
100/100
## RoGemma Model Family | Model | Link | |--------------------|:--------:| |*RoGemma-7b-Instruct-2024-06-28*| [link](https://huggingface.co/OpenLLM-Ro/RoGemma-7b-Instruct-2024-06-28) | |RoGemma-7b-Instruct-2024-10-09| [link](https://huggingface.co/OpenLLM-Ro/RoGemma-7b-Instruct-2024-10-09) | |RoGemma-7b-Instruct-DPO-2024-10-09| [link](https://huggingface.co/OpenLLM-Ro/RoGemma-7b-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}, } ```