|
--- |
|
license: apache-2.0 |
|
pipeline_tag: text-generation |
|
tags: |
|
- multilingual |
|
- PyTorch |
|
- Transformers |
|
- gpt3 |
|
- gpt2 |
|
- Deepspeed |
|
- Megatron |
|
datasets: |
|
- mc4 |
|
- Wikipedia |
|
|
|
widget: |
|
- text: "I know you're tired, but can we go for another walk this evening?\npeter szemraj:\n\n" |
|
example_title: "walk" |
|
- text: "What do you call an alligator who's just had surgery to remove his left arm?\npeter szemraj:\n\n" |
|
example_title: "alligator" |
|
- text: "If you could live anywhere, where would it be?\npeter szemraj:\n\n" |
|
example_title: "dream living place" |
|
- text: "What really makes you angry?\npeter szemraj:\n\n" |
|
example_title: "pet peeve" |
|
- text: "My friend says that she knows every language, but she doesn't speak any of them.. what's wrong with her?\npeter szemraj:\n\n" |
|
example_title: "language" |
|
- text: "What would you change about yourself if you could?\npeter szemraj:\n\n" |
|
example_title: "change" |
|
- text: "My first is in Asia, my second is in Europe, my third is in North America, and my fourth is in South America. What am I?\npeter szemraj:\n\n" |
|
example_title: "continent" |
|
- text: "Can you take me for dinner somewhere nice this time?\npeter szemraj:\n\n" |
|
example_title: "dinner" |
|
- text: "Honey, I have clogged the toilet for the third time this month.. sorry..\npeter szemraj:\n\n" |
|
example_title: "overflow" |
|
- text: "A man pushes his car to a hotel and tells the owner he's bankrupt. Why?\npeter szemraj:\n\n" |
|
example_title: "brain teaser" |
|
|
|
inference: |
|
parameters: |
|
min_length: 2 |
|
max_length: 64 |
|
length_penalty: 0.4 |
|
no_repeat_ngram_size: 3 |
|
do_sample: True |
|
top_p: 0.95 |
|
top_k: 30 |
|
temperature: 0.65 |
|
repetition_penalty: 3.5 |
|
|
|
--- |
|
|
|
|
|
# mGPT: fine-tune on message data MWE |
|
|
|
This model is a fine-tuned version of [sberbank-ai/mGPT](https://huggingface.co/sberbank-ai/mGPT) on 80k messages. Trained for one epoch, will be updated in a (separate) model repo later. |
|
|
|
## Model description |
|
|
|
- testing if fine-tuned personality data bleeds over to other languages without being trained in them explicitly |
|
|
|
### Usage in python |
|
|
|
Install the transformers library if you don't have it: |
|
|
|
``` |
|
pip install -U transformers |
|
``` |
|
|
|
load the model into a pipeline object: |
|
|
|
|
|
``` |
|
from transformers import pipeline |
|
import torch |
|
device = 'cuda' if torch.cuda.is_available() else 'cpu' |
|
my_chatbot = pipeline('text-generation', |
|
'pszemraj/mGPT-Peter-mwe', |
|
device=0 if device == 'cuda' else -1, |
|
) |
|
``` |
|
|
|
|
|
|
|
|
|
## Training procedure |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 5e-05 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 4 |
|
- seed: 42 |
|
- distributed_type: multi-GPU |
|
- gradient_accumulation_steps: 8 |
|
- total_train_batch_size: 32 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine_with_restarts |
|
- lr_scheduler_warmup_ratio: 0.05 |
|
- num_epochs: 1 |
|
|
|
### Framework versions |
|
|
|
- Transformers 4.18.0 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.12.1 |
|
|
