ruGPT-3-medium

MERA Created at 12.01.2024 14:46

0.201

The overall result

The submission does not contain all the required tasks

Ratings for leaderboard tasks

The table will scroll to the left

Task name	Result	Metric
LCS	0.102	Accuracy
RCB	0.333 / 0.167	Accuracy F1 macro
USE	0.002	Grade norm
RWSD	0.5	Accuracy
PARus	0.498	Accuracy
ruTiE	0.5	Accuracy
MultiQ	0.106 / 0.043	F1 Exact match
CheGeKa	0.005 / 0	F1 Exact match
ruModAr	0.001	Exact match
ruMultiAr	0.012	Exact match
MathLogicQA	0.248	Accuracy
ruWorldTree	0.251 / 0.248	Accuracy F1 macro
ruOpenBookQA	0.273 / 0.271	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

0.43

Accuracy

ruMMLU

0.271

Accuracy

SimpleAr

0.008

Exact match

ruHumanEval

0 / 0 / 0

Pass@k

ruHHH

0.483

ruHateSpeech

0.543

ruDetox

0.348

ruEthics

	Correct	God	Ethical
Virtue	0.076	0.03	-0.072
Law	0.083	0.035	-0.035
Moral	0.086	0.042	-0.064
Justice	0.061	0.026	-0.068
Utilitarianism	0.076	0.033	-0.063

Information about the submission:

Mera version

Torch Version

The version of the codebase

CUDA version

Precision of the model weights

Seed

Butch

Transformers version

The number of GPUs and their type

Architecture

Team:

MERA

Name of the ML model:

ruGPT-3-medium

Link to the ML model:

https://huggingface.co/ai-forever/rugpt3medium_based_on_gpt2

Additional links:

https://arxiv.org/abs/2309.10931

Architecture description:

ruGPT-3 is a Russian counterpart of GPT-3 (Brown et al., 2020). We use the model architecture description by Brown et al. and the GPT-2 code base (Radford et al., 2019) from the Transformers library. ruGPT-3 is pretrained on the language modeling objective. We use the BBPE tokenizer with the vocabulary size of 5 · 104 tokens.

Description of the training:

The model was trained with sequence length 1024 using transformers lib by the SberDevices team on 80B tokens for 3 epochs. After that, the model was finetuned 1 epoch with sequence length 2048. Total training time was around 14 days on 128 GPUs for 1024 context and a few days on 16 GPUs for 2048 context. The final perplexity on the test set is 13.6.

Pretrain data:

450GB of texts. The corpus includes texts from various publicly available resources, which represent diverse domains: Wikipedia, News, Books, Colossal Clean Crawled Corpus, OpenSubtitles.

Training Details:

The ruGPT-3 models are pretrained with a maximum sequence length of 1024 tokens for three epochs and of 2048 tokens for one epoch. We use the initial learning rate of 1e−4 and the Adam optimizer with β1 = 0.9, β2 = 0.99, and ϵ = 1e−8. The total number of tokens seen during pretraining is 80B. The pretraining of ruGPT3-large has taken 16 days on the cluster of 64 V100-SXM3 GPUs

License:

MIT

Strategy, generation and parameters:

Code version v.1.1.0 All the parameters were not changed and are used as prepared by the organizers. Details: - 1 x NVIDIA A100 - dtype auto - Pytorch 2.1.2 + CUDA 12.1 - Transformers 4.36.2 - Context length 2048

Expand information

Ratings by subcategory

Metric: Accuracy

Model, team	Honest	Helpful	Harmless
ruGPT-3-medium MERA	0.508	0.475	0.466

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
ruGPT-3-medium MERA	0.4	0.25	0.2	0.314	0.19	0.3	0.4	0.118	0.4	0.2	0.455	0.7	0.3	0.385	0.182	0.1	0.2	0.333	0.6	0.2	0.2	0.288	0.1	0.196	0.3	0.273	0.188	0.143	0.2	0.273	0.2	0.444	0.6	0.2	0.182	0.3	0.1	0.095	0.3	0.2	0.228	0.2	0.4	0.4	0.5	0.1	0.4	0.3	0.4	0	0.227	0.188	0.324	0.2	0.167	0.333	0.333

Model, team	SIM	FL	STA
ruGPT-3-medium MERA	0.713	0.618	0.755

Coorect

Good

Ethical

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
ruGPT-3-medium MERA	0.076	0.083	0.086	0.061	0.076

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
ruGPT-3-medium MERA	0.03	0.035	0.042	0.026	0.033

Model, team	Virtue	Law	Moral	Justice	Utilitarianism
ruGPT-3-medium MERA	-0.072	-0.035	-0.064	-0.068	-0.063

Model, team	Women	Men	LGBT	Nationalities	Migrants	Other
ruGPT-3-medium MERA	0.519	0.686	0.588	0.595	0.286	0.492

ruGPT-3-medium

Ratings for leaderboard tasks

Evaluation on open tasks:

Information about the submission:

Team:

Name of the ML model:

Link to the ML model:

Additional links:

Architecture description:

Description of the training:

Pretrain data:

Training Details:

License:

Strategy, generation and parameters:

Ratings by subcategory

Confirm the deletion of the sub