metadata

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
  - OpenLLM-Ro/ro_sft_oasst
  - OpenLLM-Ro/ro_sft_ultrachat
model-index:
  - name: OpenLLM-Ro/RoGemma-7b-Instruct-2024-10-09
    results:
      - task:
          type: text-generation
        dataset:
          name: RoMT-Bench
          type: RoMT-Bench
        metrics:
          - name: Score
            type: Score
            value: 5.24
      - task:
          type: text-generation
        dataset:
          name: RoCulturaBench
          type: RoCulturaBench
        metrics:
          - name: Score
            type: Score
            value: 3.51
      - task:
          type: text-generation
        dataset:
          name: Romanian_Academic_Benchmarks
          type: Romanian_Academic_Benchmarks
        metrics:
          - name: Average accuracy
            type: accuracy
            value: 50.48
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_arc_challenge
          type: OpenLLM-Ro/ro_arc_challenge
        metrics:
          - name: Average accuracy
            type: accuracy
            value: 52.01
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_mmlu
          type: OpenLLM-Ro/ro_mmlu
        metrics:
          - name: Average accuracy
            type: accuracy
            value: 52.37
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_winogrande
          type: OpenLLM-Ro/ro_winogrande
        metrics:
          - name: Average accuracy
            type: accuracy
            value: 66.97
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_hellaswag
          type: OpenLLM-Ro/ro_hellaswag
        metrics:
          - name: Average accuracy
            type: accuracy
            value: 56.34
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_gsm8k
          type: OpenLLM-Ro/ro_gsm8k
        metrics:
          - name: Average accuracy
            type: accuracy
            value: 25.98
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_truthfulqa
          type: OpenLLM-Ro/ro_truthfulqa
        metrics:
          - name: Average accuracy
            type: accuracy
            value: 49.18
      - task:
          type: text-generation
        dataset:
          name: LaRoSeDa_binary
          type: LaRoSeDa_binary
        metrics:
          - name: Average macro-f1
            type: macro-f1
            value: 86.96
      - task:
          type: text-generation
        dataset:
          name: LaRoSeDa_multiclass
          type: LaRoSeDa_multiclass
        metrics:
          - name: Average macro-f1
            type: macro-f1
            value: 56.72
      - task:
          type: text-generation
        dataset:
          name: LaRoSeDa_binary_finetuned
          type: LaRoSeDa_binary_finetuned
        metrics:
          - name: Average macro-f1
            type: macro-f1
            value: 98.8
      - task:
          type: text-generation
        dataset:
          name: LaRoSeDa_multiclass_finetuned
          type: LaRoSeDa_multiclass_finetuned
        metrics:
          - name: Average macro-f1
            type: macro-f1
            value: 85.81
      - task:
          type: text-generation
        dataset:
          name: WMT_EN-RO
          type: WMT_EN-RO
        metrics:
          - name: Average bleu
            type: bleu
            value: 24.45
      - task:
          type: text-generation
        dataset:
          name: WMT_RO-EN
          type: WMT_RO-EN
        metrics:
          - name: Average bleu
            type: bleu
            value: 14.2
      - task:
          type: text-generation
        dataset:
          name: WMT_EN-RO_finetuned
          type: WMT_EN-RO_finetuned
        metrics:
          - name: Average bleu
            type: bleu
            value: 25.96
      - task:
          type: text-generation
        dataset:
          name: WMT_RO-EN_finetuned
          type: WMT_RO-EN_finetuned
        metrics:
          - name: Average bleu
            type: bleu
            value: 39.07
      - task:
          type: text-generation
        dataset:
          name: XQuAD
          type: XQuAD
        metrics:
          - name: Average exact_match
            type: exact_match
            value: 26.03
      - task:
          type: text-generation
        dataset:
          name: XQuAD
          type: XQuAD
        metrics:
          - name: Average f1
            type: f1
            value: 41.58
      - task:
          type: text-generation
        dataset:
          name: XQuAD_finetuned
          type: XQuAD_finetuned
        metrics:
          - name: Average exact_match
            type: exact_match
            value: 46.72
      - task:
          type: text-generation
        dataset:
          name: XQuAD_finetuned
          type: XQuAD_finetuned
        metrics:
          - name: Average f1
            type: f1
            value: 60.79
      - task:
          type: text-generation
        dataset:
          name: STS
          type: STS
        metrics:
          - name: Average spearman
            type: spearman
            value: 73.23
      - task:
          type: text-generation
        dataset:
          name: STS
          type: STS
        metrics:
          - name: Average pearson
            type: pearson
            value: 71.58
      - task:
          type: text-generation
        dataset:
          name: STS_finetuned
          type: STS_finetuned
        metrics:
          - name: Average spearman
            type: spearman
            value: 88.42
      - task:
          type: text-generation
        dataset:
          name: STS_finetuned
          type: STS_finetuned
        metrics:
          - name: Average pearson
            type: pearson
            value: 88.45
      - task:
          type: text-generation
        dataset:
          name: RoMT-Bench
          type: RoMT-Bench
        metrics:
          - name: First turn
            type: Score
            value: 5.55
          - name: Second turn
            type: Score
            value: 4.94
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_arc_challenge
          type: OpenLLM-Ro/ro_arc_challenge
        metrics:
          - name: 0-shot
            type: accuracy
            value: 49.53
          - name: 1-shot
            type: accuracy
            value: 52.53
          - name: 3-shot
            type: accuracy
            value: 51.5
          - name: 5-shot
            type: accuracy
            value: 53.56
          - name: 10-shot
            type: accuracy
            value: 52.53
          - name: 25-shot
            type: accuracy
            value: 52.44
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_mmlu
          type: OpenLLM-Ro/ro_mmlu
        metrics:
          - name: 0-shot
            type: accuracy
            value: 51.81
          - name: 1-shot
            type: accuracy
            value: 52.45
          - name: 3-shot
            type: accuracy
            value: 52.52
          - name: 5-shot
            type: accuracy
            value: 52.7
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_winogrande
          type: OpenLLM-Ro/ro_winogrande
        metrics:
          - name: 0-shot
            type: accuracy
            value: 66.54
          - name: 1-shot
            type: accuracy
            value: 66.69
          - name: 3-shot
            type: accuracy
            value: 67.09
          - name: 5-shot
            type: accuracy
            value: 67.56
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_hellaswag
          type: OpenLLM-Ro/ro_hellaswag
        metrics:
          - name: 0-shot
            type: accuracy
            value: 58.8
          - name: 1-shot
            type: accuracy
            value: 57.04
          - name: 3-shot
            type: accuracy
            value: 55.85
          - name: 5-shot
            type: accuracy
            value: 54.15
          - name: 10-shot
            type: accuracy
            value: 55.88
      - task:
          type: text-generation
        dataset:
          name: OpenLLM-Ro/ro_gsm8k
          type: OpenLLM-Ro/ro_gsm8k
        metrics:
          - name: 1-shot
            type: accuracy
            value: 22.06
          - name: 3-shot
            type: accuracy
            value: 25.4
          - name: 5-shot
            type: accuracy
            value: 30.48
      - task:
          type: text-generation
        dataset:
          name: LaRoSeDa_binary
          type: LaRoSeDa_binary
        metrics:
          - name: 0-shot
            type: macro-f1
            value: 87.28
          - name: 1-shot
            type: macro-f1
            value: 86.4
          - name: 3-shot
            type: macro-f1
            value: 87.95
          - name: 5-shot
            type: macro-f1
            value: 86.2
      - task:
          type: text-generation
        dataset:
          name: LaRoSeDa_multiclass
          type: LaRoSeDa_multiclass
        metrics:
          - name: 0-shot
            type: macro-f1
            value: 38.35
          - name: 1-shot
            type: macro-f1
            value: 63.86
          - name: 3-shot
            type: macro-f1
            value: 62.03
          - name: 5-shot
            type: macro-f1
            value: 62.62
      - task:
          type: text-generation
        dataset:
          name: WMT_EN-RO
          type: WMT_EN-RO
        metrics:
          - name: 0-shot
            type: bleu
            value: 11.39
          - name: 1-shot
            type: bleu
            value: 28.08
          - name: 3-shot
            type: bleu
            value: 29.18
          - name: 5-shot
            type: bleu
            value: 29.13
      - task:
          type: text-generation
        dataset:
          name: WMT_RO-EN
          type: WMT_RO-EN
        metrics:
          - name: 0-shot
            type: bleu
            value: 1.92
          - name: 1-shot
            type: bleu
            value: 9.39
          - name: 3-shot
            type: bleu
            value: 21.81
          - name: 5-shot
            type: bleu
            value: 23.66
      - task:
          type: text-generation
        dataset:
          name: XQuAD_EM
          type: XQuAD_EM
        metrics:
          - name: 0-shot
            type: exact_match
            value: 32.77
          - name: 1-shot
            type: exact_match
            value: 20.25
          - name: 3-shot
            type: exact_match
            value: 18.49
          - name: 5-shot
            type: exact_match
            value: 32.6
      - task:
          type: text-generation
        dataset:
          name: XQuAD_F1
          type: XQuAD_F1
        metrics:
          - name: 0-shot
            type: f1
            value: 47.98
          - name: 1-shot
            type: f1
            value: 34.92
          - name: 3-shot
            type: f1
            value: 33.27
          - name: 5-shot
            type: f1
            value: 50.14
      - task:
          type: text-generation
        dataset:
          name: STS_Spearman
          type: STS_Spearman
        metrics:
          - name: 1-shot
            type: spearman
            value: 71.75
          - name: 3-shot
            type: spearman
            value: 71.83
          - name: 5-shot
            type: spearman
            value: 76.11
      - task:
          type: text-generation
        dataset:
          name: STS_Pearson
          type: STS_Pearson
        metrics:
          - name: 1-shot
            type: pearson
            value: 69.97
          - name: 3-shot
            type: pearson
            value: 69.87
          - name: 5-shot
            type: pearson
            value: 74.89

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
Trained using: RoAlpaca, RoAlpacaGPT4, RoDolly, RoSelfInstruct, RoNoRobots, RoOrca, RoCamel, RoOpenAssistant, RoUltraChat

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.

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("OpenLLM-Ro/RoGemma-7b-Instruct-2024-10-09")
model = AutoModelForCausalLM.from_pretrained("OpenLLM-Ro/RoGemma-7b-Instruct-2024-10-09")

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
RoGemma-7b-Instruct-2024-10-09	link
RoGemma-7b-Instruct-DPO-2024-10-09	link

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}, 
}