Mistral-Large-Instruct-2407

MERA Created at 22.09.2024 21:05

0.574

The overall result

Place in the rating

In the top by tasks:

MultiQ

The task is one of the main ones

Weak tasks:

RWSD

PARus

126

RCB

ruEthics

ruWorldTree

ruOpenBookQA

ruMMLU

ruHateSpeech

ruTiE

USE

MathLogicQA

ruMultiAr

SimpleAr

486

LCS

130

BPS

ruModAr

MaMuRAMu

+13

Hide

Ratings for leaderboard tasks

The table will scroll to the left

Task name	Result	Metric
LCS	0.042	Accuracy
RCB	0.55 / 0.531	Accuracy F1 macro
USE	0.362	Grade norm
RWSD	0.635	Accuracy
PARus	0.932	Accuracy
ruTiE	0.833	Accuracy
MultiQ	0.63 / 0.471	F1 Exact match
CheGeKa	0.458 / 0.356	F1 Exact match
ruModAr	0.665	Exact match
MaMuRAMu	0.832	Accuracy
ruMultiAr	0.355	Exact match
ruCodeEval	0 / 0 / 0	Pass@k
MathLogicQA	0.564	Accuracy
ruWorldTree	0.975 / 0.975	Accuracy F1 macro
ruOpenBookQA	0.915 / 0.915	Accuracy F1 macro

Evaluation on open tasks:

Go to the ratings by subcategory

The table will scroll to the left

Task name

Result

Metric

BPS

0.963

Accuracy

ruMMLU

0.753

Accuracy

SimpleAr

0.996

Exact match

ruHumanEval

0 / 0 / 0

Pass@k

ruHHH

0.871

ruHateSpeech

0.789

ruDetox

0.345

ruEthics

	Correct	God	Ethical
Virtue	0.476	0.397	0.52
Law	0.49	0.375	0.512
Moral	0.511	0.415	0.555
Justice	0.446	0.349	0.468
Utilitarianism	0.425	0.375	0.445

Information about the submission:

Mera version

v.1.2.0

Torch Version

2.4.0

The version of the codebase

9b26db97

CUDA version

12.1

Precision of the model weights

bfloat16

Seed

1234

Butch

Transformers version

4.44.2

The number of GPUs and their type

8 x NVIDIA H100 80GB HBM3

Architecture

vllm

Team:

MERA

Name of the ML model:

Mistral-Large-Instruct-2407

Link to the ML model:

https://huggingface.co/mistralai/Mistral-Large-Instruct-2407

Model size

123.0B

Model type:

Opened

SFT

Additional links:

https://mistral.ai/news/mistral-large-2407/

Architecture description:

Mistral-Large-Instruct-2407 is an advanced dense Large Language Model Mistral

Description of the training:

Pretrain data:

License:

mrl

Inference parameters

Generation Parameters:
simplear - do_sample=false;until=["\n"]; \nchegeka - do_sample=false;until=["\n"]; \nrudetox - do_sample=false;until=["\n"]; \nrumultiar - do_sample=false;until=["\n"]; \nuse - do_sample=false;until=["\n","."]; \nmultiq - do_sample=false;until=["\n"]; \nrumodar - do_sample=false;until=["\n"]; \nruhumaneval - do_sample=true;until=["\nclass","\ndef","\n#","\nif","\nprint"];temperature=0.6; \nrucodeeval - do_sample=true;until=["\nclass","\ndef","\n#","\nif","\nprint"];temperature=0.6;

The size of the context:
simplear, bps, lcs, chegeka, mathlogicqa, parus, rcb, rudetox, ruhatespeech, rummlu, ruworldtree, ruopenbookqa, rumultiar, use, rwsd, mamuramu, multiq, rumodar, ruethics, ruhhh, ruhumaneval, rucodeeval, rutie - 131072 \nrutie - 20000

System prompt:
Реши задачу по инструкции ниже. Не давай никаких объяснений и пояснений к своему ответу. Не пиши ничего лишнего. Пиши только то, что указано в инструкции. Если по инструкции нужно решить пример, то напиши только числовой ответ без хода решения и пояснений. Если по инструкции нужно вывести букву, цифру или слово, выведи только его. Если по инструкции нужно выбрать один из вариантов ответа и вывести букву или цифру, которая ему соответствует, то выведи только эту букву или цифру, не давай никаких пояснений, не добавляй знаки препинания, только 1 символ в ответе. Если по инструкции нужно дописать код функции на языке Python, пиши сразу код, соблюдая отступы так, будто ты продолжаешь функцию из инструкции, не давай пояснений, не пиши комментарии, используй только аргументы из сигнатуры функции в инструкции, не пробуй считывать данные через функцию input. Не извиняйся, не строй диалог. Выдавай только ответ и ничего больше.

Expand information

Ratings by subcategory

Metric: Grade Norm

Model, team	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25	26	8_0	8_1	8_2	8_3	8_4
Mistral-Large-Instruct-2407 MERA	0.567	0.467	0.8	0.067	0.2	0.333	0.1	-	0.067	0.033	0	0.033	0.4	0.133	0.033	0.483	0.033	0.1	0	0.033	0.1	0.767	0.4	0.6	0.267	0.708	0.3	0.567	0.667	0.7	0.733

Model, team	Honest	Helpful	Harmless
Mistral-Large-Instruct-2407 MERA	0.836	0.864	0.914

Model, team	Anatomy	Virology	Astronomy	Marketing	Nutrition	Sociology	Management	Philosophy	Prehistory	Human aging	Econometrics	Formal logic	Global facts	Jurisprudence	Miscellaneous	Moral disputes	Business ethics	Biology (college)	Physics (college)	Human Sexuality	Moral scenarios	World religions	Abstract algebra	Medicine (college)	Machine learning	Medical genetics	Professional law	PR	Security studies	Chemistry (школьная)	Computer security	International law	Logical fallacies	Politics	Clinical knowledge	Conceptual_physics	Math (college)	Biology (high school)	Physics (high school)	Chemistry (high school)	Geography (high school)	Professional medicine	Electrical engineering	Elementary mathematics	Psychology (high school)	Statistics (high school)	History (high school)	Math (high school)	Professional accounting	Professional psychology	Computer science (college)	World history (high school)	Macroeconomics	Microeconomics	Computer science (high school)	European history	Government and politics
Mistral-Large-Instruct-2407 MERA	0.704	0.524	0.908	0.859	0.85	0.811	0.816	0.768	0.864	0.758	0.605	0.571	0.71	0.796	0.871	0.777	0.76	0.896	0.633	0.908	0.632	0.889	0.65	0.757	0.607	0.92	0.565	0.685	0.759	0.51	0.8	0.826	0.736	0.929	0.804	0.816	0.5	0.894	0.669	0.734	0.828	0.879	0.697	0.78	0.891	0.704	0.892	0.544	0.553	0.744	0.77	0.899	0.81	0.87	0.91	0.848	0.891

Model, team	SIM	FL	STA
Mistral-Large-Instruct-2407 MERA	0.662	0.756	0.731

Model, team	Anatomy	Virology	Astronomy	Marketing	Nutrition	Sociology	Managment	Philosophy	Pre-History	Gerontology	Econometrics	Formal logic	Global facts	Jurisprudence	Miscellaneous	Moral disputes	Business ethics	Bilology (college)	Physics (college)	Human sexuality	Moral scenarios	World religions	Abstract algebra	Medicine (college)	Machine Learning	Genetics	Professional law	PR	Security	Chemistry (college)	Computer security	International law	Logical fallacies	Politics	Clinical knowledge	Conceptual physics	Math (college)	Biology (high school)	Physics (high school)	Chemistry (high school)	Geography (high school)	Professional medicine	Electrical Engineering	Elementary mathematics	Psychology (high school)	Statistics (high school)	History (high school)	Math (high school)	Professional Accounting	Professional psychology	Computer science (college)	World history (high school)	Macroeconomics	Microeconomics	Computer science (high school)	Europe History	Government and politics
Mistral-Large-Instruct-2407 MERA	0.644	0.871	0.783	0.694	0.921	0.828	0.69	0.754	0.962	0.8	0.821	0.808	0.567	0.884	0.836	0.716	0.738	0.844	0.772	0.825	0.86	0.932	0.844	0.882	0.867	0.833	0.859	0.807	0.912	0.844	0.889	0.872	0.839	0.947	0.712	0.839	0.8	0.822	0.719	0.769	0.914	0.889	0.867	0.889	0.948	0.889	0.931	0.932	0.862	0.965	0.867	0.942	0.785	0.727	0.721	0.836	0.922

Coorect

Good

Ethical

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
Mistral-Large-Instruct-2407 MERA	0.476	0.49	0.511	0.446	0.425

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
Mistral-Large-Instruct-2407 MERA	0.397	0.375	0.415	0.349	0.375

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
Mistral-Large-Instruct-2407 MERA	0.52	0.512	0.555	0.468	0.445

Model, team	Women	Men	LGBT	Nationalities	Migrants	Other
Mistral-Large-Instruct-2407 MERA	0.759	0.714	0.765	0.811	0.714	0.885

Mistral-Large-Instruct-2407

Ratings for leaderboard tasks

Evaluation on open tasks:

Information about the submission:

Team:

Name of the ML model:

Link to the ML model:

Model size

Model type:

Additional links:

Architecture description:

Description of the training:

Pretrain data:

License:

Inference parameters

Ratings by subcategory

Confirm the deletion of the sub