Instructions ruGPT Medium v0.11_75к_a

Model Summary

Это ruGPTMedium дообученная в инструктивно-флановом сетапе, она более ли менее ZSшотиться и FSшотиться и работает лучше чем XGLM1.7b, mgpt на русском языке

Quick Start

from transformers import pipeline
#в душе не ебу будет ли норм работать, ставлю жопу автора хф что токенайзер мисматчнет с моделью, вообще грузите по нормальному
pipe = pipeline(model='AlexWortega/instruct_rugptMedium')
pipe('''Как собрать питон код?''')

or

from transformers import GPT2TokenizerFast,GPT2LMHeadModel
tokenizer = GPT2TokenizerFast.from_pretrained("AlexWortega/instruct_rugptMedium")
special_tokens_dict = {'additional_special_tokens': ['<code>', '</code>', '<instructionS>', '<instructionE>', '<next>']}

tokenizer.add_special_tokens(special_tokens_dict)
device = 'cuda:1'
model = GPT2LMHeadModel.from_pretrained("AlexWortega/instruct_rugptMedium")
model.to(device)

model.resize_token_embeddings(len(tokenizer))

обратите внимание, что лучшие параметры для генерации

gen_kwargs = {
        "min_length": 20,
        "max_new_tokens": 100,
        "top_k": 50,
        "top_p": 0.9,
        "do_sample": True,  
        "early_stopping": True,
        "no_repeat_ngram_size": 2,
        "eos_token_id": tokenizer.eos_token_id,
        "pad_token_id": tokenizer.eos_token_id,
        "use_cache": True,
        "repetition_penalty": 1.5,  
        "length_penalty": 0.8,  
        "num_beams": 4,
        "num_return_sequences": k
    }

License

The weights of Instructions ruGPT Small v0.1a are licensed under version 2.0 of the Apache License.

Hyperparameters

I used Novograd with a learning rate of 2e-5 and global batch size of 6 (3 for each data parallel worker). I use both data parallelism and pipeline parallelism to conduct training. During training, we truncate the input sequence to 1024 tokens, and for input sequence that contains less than 1024 tokens, we concatenate multiple sequences into one long sequence to improve the data efficiency.

References

#Metrics

SOON

BibTeX entry and citation info

@article{
  title={GPT2xl is underrated task solver},
  author={Nickolich Aleksandr, 5Q, datascience, Ilya Gusev, Alex Kukushkin, Karina Romanova, Arseniy Shahmatov, Maksim Gersimenko},
  year={2023}
}