YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

Quantization made by Richard Erkhov.

Request more models

RoLlama2-7b-Instruct-2024-05-14 - GGUF

Model creator: https://huggingface.co/OpenLLM-Ro/
Original model: https://huggingface.co/OpenLLM-Ro/RoLlama2-7b-Instruct-2024-05-14/

Name	Quant method	Size
RoLlama2-7b-Instruct-2024-05-14.Q2_K.gguf	Q2_K	2.36GB
RoLlama2-7b-Instruct-2024-05-14.IQ3_XS.gguf	IQ3_XS	2.6GB
RoLlama2-7b-Instruct-2024-05-14.IQ3_S.gguf	IQ3_S	2.75GB
RoLlama2-7b-Instruct-2024-05-14.Q3_K_S.gguf	Q3_K_S	2.75GB
RoLlama2-7b-Instruct-2024-05-14.IQ3_M.gguf	IQ3_M	2.9GB
RoLlama2-7b-Instruct-2024-05-14.Q3_K.gguf	Q3_K	3.07GB
RoLlama2-7b-Instruct-2024-05-14.Q3_K_M.gguf	Q3_K_M	3.07GB
RoLlama2-7b-Instruct-2024-05-14.Q3_K_L.gguf	Q3_K_L	3.35GB
RoLlama2-7b-Instruct-2024-05-14.IQ4_XS.gguf	IQ4_XS	3.4GB
RoLlama2-7b-Instruct-2024-05-14.Q4_0.gguf	Q4_0	3.56GB
RoLlama2-7b-Instruct-2024-05-14.IQ4_NL.gguf	IQ4_NL	3.58GB
RoLlama2-7b-Instruct-2024-05-14.Q4_K_S.gguf	Q4_K_S	3.59GB
RoLlama2-7b-Instruct-2024-05-14.Q4_K.gguf	Q4_K	3.8GB
RoLlama2-7b-Instruct-2024-05-14.Q4_K_M.gguf	Q4_K_M	3.8GB
RoLlama2-7b-Instruct-2024-05-14.Q4_1.gguf	Q4_1	3.95GB
RoLlama2-7b-Instruct-2024-05-14.Q5_0.gguf	Q5_0	4.33GB
RoLlama2-7b-Instruct-2024-05-14.Q5_K_S.gguf	Q5_K_S	4.33GB
RoLlama2-7b-Instruct-2024-05-14.Q5_K.gguf	Q5_K	4.45GB
RoLlama2-7b-Instruct-2024-05-14.Q5_K_M.gguf	Q5_K_M	4.45GB
RoLlama2-7b-Instruct-2024-05-14.Q5_1.gguf	Q5_1	4.72GB
RoLlama2-7b-Instruct-2024-05-14.Q6_K.gguf	Q6_K	5.15GB
RoLlama2-7b-Instruct-2024-05-14.Q8_0.gguf	Q8_0	6.67GB

Original model description:

license: cc-by-nc-4.0 language: - ro base_model: - OpenLLM-Ro/RoLlama2-7b-Base model-index: - name: OpenLLM-Ro/RoLlama2-7b-Instruct-2024-05-14 results: - task: type: text-generation dataset: name: RoMT-Bench type: RoMT-Bench metrics: - name: Score type: Score value: 3.86 - task: type: text-generation dataset: name: RoCulturaBench type: RoCulturaBench metrics: - name: Score type: Score value: 3.77 - task: type: text-generation dataset: name: Romanian_Academic_Benchmarks type: Romanian_Academic_Benchmarks metrics: - name: Average accuracy type: accuracy value: 45.71 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_arc_challenge type: OpenLLM-Ro/ro_arc_challenge metrics: - name: Average accuracy type: accuracy value: 43.66 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_mmlu type: OpenLLM-Ro/ro_mmlu metrics: - name: Average accuracy type: accuracy value: 39.7 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_winogrande type: OpenLLM-Ro/ro_winogrande metrics: - name: Average accuracy type: accuracy value: 70.34 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_hellaswag type: OpenLLM-Ro/ro_hellaswag metrics: - name: Average accuracy type: accuracy value: 57.36 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_gsm8k type: OpenLLM-Ro/ro_gsm8k metrics: - name: Average accuracy type: accuracy value: 18.78 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_truthfulqa type: OpenLLM-Ro/ro_truthfulqa metrics: - name: Average accuracy type: accuracy value: 44.44 - task: type: text-generation dataset: name: LaRoSeDa_binary type: LaRoSeDa_binary metrics: - name: Average macro-f1 type: macro-f1 value: 97.48 - task: type: text-generation dataset: name: LaRoSeDa_multiclass type: LaRoSeDa_multiclass metrics: - name: Average macro-f1 type: macro-f1 value: 65.26 - task: type: text-generation dataset: name: LaRoSeDa_binary_finetuned type: LaRoSeDa_binary_finetuned metrics: - name: Average macro-f1 type: macro-f1 value: 98.83 - task: type: text-generation dataset: name: LaRoSeDa_multiclass_finetuned type: LaRoSeDa_multiclass_finetuned metrics: - name: Average macro-f1 type: macro-f1 value: 87.28 - task: type: text-generation dataset: name: WMT_EN-RO type: WMT_EN-RO metrics: - name: Average bleu type: bleu value: 27.38 - task: type: text-generation dataset: name: WMT_RO-EN type: WMT_RO-EN metrics: - name: Average bleu type: bleu value: 10.32 - task: type: text-generation dataset: name: WMT_EN-RO_finetuned type: WMT_EN-RO_finetuned metrics: - name: Average bleu type: bleu value: 27.59 - task: type: text-generation dataset: name: WMT_RO-EN_finetuned type: WMT_RO-EN_finetuned metrics: - name: Average bleu type: bleu value: 40.13 - task: type: text-generation dataset: name: XQuAD type: XQuAD metrics: - name: Average exact_match type: exact_match value: 44.52 - task: type: text-generation dataset: name: XQuAD type: XQuAD metrics: - name: Average f1 type: f1 value: 64.75 - task: type: text-generation dataset: name: XQuAD_finetuned type: XQuAD_finetuned metrics: - name: Average exact_match type: exact_match value: 54.96 - task: type: text-generation dataset: name: XQuAD_finetuned type: XQuAD_finetuned metrics: - name: Average f1 type: f1 value: 70.2 - task: type: text-generation dataset: name: STS type: STS metrics: - name: Average spearman type: spearman value: 65.5 - task: type: text-generation dataset: name: STS type: STS metrics: - name: Average pearson type: pearson value: 67.79 - task: type: text-generation dataset: name: STS_finetuned type: STS_finetuned metrics: - name: Average spearman type: spearman value: 84.44 - task: type: text-generation dataset: name: STS_finetuned type: STS_finetuned metrics: - name: Average pearson type: pearson value: 84.76 - task: type: text-generation dataset: name: RoMT-Bench type: RoMT-Bench metrics: - name: First turn type: Score value: 4.67 - name: Second turn type: Score value: 3.04 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_arc_challenge type: OpenLLM-Ro/ro_arc_challenge metrics: - name: 0-shot type: accuracy value: 41.73 - name: 1-shot type: accuracy value: 42.16 - name: 3-shot type: accuracy value: 43.53 - name: 5-shot type: accuracy value: 44.9 - name: 10-shot type: accuracy value: 44.99 - name: 25-shot type: accuracy value: 44.64 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_mmlu type: OpenLLM-Ro/ro_mmlu metrics: - name: 0-shot type: accuracy value: 38.54 - name: 1-shot type: accuracy value: 39.36 - name: 3-shot type: accuracy value: 40.82 - name: 5-shot type: accuracy value: 40.07 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_winogrande type: OpenLLM-Ro/ro_winogrande metrics: - name: 0-shot type: accuracy value: 72.61 - name: 1-shot type: accuracy value: 69.93 - name: 3-shot type: accuracy value: 70.4 - name: 5-shot type: accuracy value: 68.43 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_hellaswag type: OpenLLM-Ro/ro_hellaswag metrics: - name: 0-shot type: accuracy value: 56.9 - name: 1-shot type: accuracy value: 57.07 - name: 3-shot type: accuracy value: 57.56 - name: 5-shot type: accuracy value: 57.35 - name: 10-shot type: accuracy value: 57.93 - task: type: text-generation dataset: name: OpenLLM-Ro/ro_gsm8k type: OpenLLM-Ro/ro_gsm8k metrics: - name: 0-shot type: accuracy value: 11.22 - name: 1-shot type: accuracy value: 21.38 - name: 3-shot type: accuracy value: 23.73 - task: type: text-generation dataset: name: LaRoSeDa_binary type: LaRoSeDa_binary metrics: - name: 0-shot type: macro-f1 value: 97.67 - name: 1-shot type: macro-f1 value: 96.77 - name: 3-shot type: macro-f1 value: 97.6 - name: 5-shot type: macro-f1 value: 97.87 - task: type: text-generation dataset: name: LaRoSeDa_multiclass type: LaRoSeDa_multiclass metrics: - name: 0-shot type: macro-f1 value: 61.82 - name: 1-shot type: macro-f1 value: 58.84 - name: 3-shot type: macro-f1 value: 68.67 - name: 5-shot type: macro-f1 value: 71.71 - task: type: text-generation dataset: name: WMT_EN-RO type: WMT_EN-RO metrics: - name: 0-shot type: bleu value: 19.71 - name: 1-shot type: bleu value: 29.62 - name: 3-shot type: bleu value: 30.11 - name: 5-shot type: bleu value: 30.1 - task: type: text-generation dataset: name: WMT_RO-EN type: WMT_RO-EN metrics: - name: 0-shot type: bleu value: 1.86 - name: 1-shot type: bleu value: 4.41 - name: 3-shot type: bleu value: 14.95 - name: 5-shot type: bleu value: 20.07 - task: type: text-generation dataset: name: XQuAD_EM type: XQuAD_EM metrics: - name: 0-shot type: exact_match value: 34.87 - name: 1-shot type: exact_match value: 44.96 - name: 3-shot type: exact_match value: 48.4 - name: 5-shot type: exact_match value: 49.83 - task: type: text-generation dataset: name: XQuAD_F1 type: XQuAD_F1 metrics: - name: 0-shot type: f1 value: 58.07 - name: 1-shot type: f1 value: 63.93 - name: 3-shot type: f1 value: 67.89 - name: 5-shot type: f1 value: 69.1 - task: type: text-generation dataset: name: STS type: STS metrics: - name: 0-shot type: spearman value: 61.14 - name: 1-shot type: spearman value: 66.91 - name: 3-shot type: spearman value: 68.46 - task: type: text-generation dataset: name: STS type: STS metrics: - name: 0-shot type: pearson value: 61.88 - name: 1-shot type: pearson value: 70.04 - name: 3-shot type: pearson value: 71.46 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 Card for Model ID

RoLlama2 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 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: RoLlama2-7b-Base
Trained using: RoAlpaca, RoAlpacaGPT4, RoDolly, RoSelfInstruct, RoNoRobots, RoOrca, RoCamel

Model Sources

Repository: https://github.com/OpenLLM-Ro/llama-recipes
Paper: https://arxiv.org/abs/2406.18266

Intended Use

Intended Use Cases

RoLlama2 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/RoLlama2-7b-Instruct-2024-05-14 ")
model = AutoModelForCausalLM.from_pretrained("OpenLLM-Ro/RoLlama2-7b-Instruct-2024-05-14 ")

instruction = "Care este cel mai înalt vârf muntos din România?"
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)

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-2-7b-chat	36.84	37.03	33.80	55.87	45.36	4.90	44.09
RoLlama2-7b-Instruct-2024-05-14	45.71	43.66	39.70	70.34	57.36	18.78	44.44
RoLlama2-7b-Instruct-2024-10-09	44.50	44.73	40.39	63.67	59.12	13.29	45.78
RoLlama2-7b-Instruct-DPO-2024-10-09	43.20	44.24	38.39	62.57	59.20	15.72	39.07

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-2-7b-chat	87.78	52.81	97.27	82.02	15.55	28.53	19.99	31.48
RoLlama2-7b-Instruct-2024-05-14	97.48	65.26	98.83	87.28	27.38	10.32	27.59	40.13
RoLlama2-7b-Instruct-2024-10-09	97.66	62.41	97.97	60.89	27.13	19.39	27.63	39.75
RoLlama2-7b-Instruct-DPO-2024-10-09	97.31	60.56	-	-	26.56	21.68	-	-

	XQuAD				STS
	Few-shot		Finetuned		Few-shot		Finetuned
Model	(EM)	(F1)	(EM)	(F1)	(Spearman)	(Pearson)	(Spearman)	(Pearson)
Llama-2-7b-chat	32.35	54.00	60.34	75.98	32.56	31.99	74.08	72.64
RoLlama2-7b-Instruct-2024-05-14	44.52	64.75	54.96	70.20	65.50	67.79	84.44	84.76
RoLlama2-7b-Instruct-2024-10-09	45.71	65.08	59.24	74.25	59.69	57.16	84.66	85.07
RoLlama2-7b-Instruct-DPO-2024-10-09	35.78	59.31	-	-	61.22	58.41	-	-

Romanian MT-Bench

Model	Average	1st turn	2nd turn	Answers in Ro
Llama-2-7b-chat	1.08	1.44	0.73	45/160
RoLlama2-7b-Instruct-2024-05-14	3.86	4.67	3.04	160/160
RoLlama2-7b-Instruct-2024-10-09	4.43	4.92	3.94	160/160
RoLlama2-7b-Instruct-DPO-2024-10-09	4.61	5.15	4.06	160/160

RoCulturaBench

Model	Average	Answers in Ro
Llama-2-7b-chat	1.21	33/100
RoLlama2-7b-Instruct-2024-05-14	3.77	100/100
RoLlama2-7b-Instruct-2024-10-09	4.08	100/100
RoLlama2-7b-Instruct-DPO-2024-10-09	4.80	100/100

RoLlama2 Model Family

Model	Link
RoLlama2-7b-Base-2024-05-14	link
RoLlama2-7b-Instruct-2024-05-14	link
RoLlama2-7b-Instruct-2024-10-09	link
RoLlama2-7b-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}, 
}

Downloads last month: 648

GGUF

Model size

7B params

Architecture

llama

Hardware compatibility

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support