|
|
--- |
|
|
license: apache-2.0 |
|
|
language: |
|
|
- en |
|
|
tags: |
|
|
- ctranslate2 |
|
|
- int8 |
|
|
- float16 |
|
|
- sft |
|
|
pipeline_tag: text-generation |
|
|
widget: |
|
|
- text: <|prompter|>What is a meme, and what's the history behind this word?<|endoftext|><|assistant|> |
|
|
- text: <|prompter|>What's the Earth total population<|endoftext|><|assistant|> |
|
|
- text: <|prompter|>Write a story about future of AI development<|endoftext|><|assistant|> |
|
|
--- |
|
|
# # Fast-Inference with Ctranslate2 |
|
|
Speedup inference while reducing memory by 2x-4x using int8 inference in C++ on CPU or GPU. |
|
|
|
|
|
quantized version of [OpenAssistant/pythia-12b-sft-v8-7k-steps](https://huggingface.co/OpenAssistant/pythia-12b-sft-v8-7k-steps) |
|
|
```bash |
|
|
pip install hf-hub-ctranslate2>=2.0.6 |
|
|
``` |
|
|
Converted on 2023-05-19 using |
|
|
``` |
|
|
ct2-transformers-converter --model OpenAssistant/pythia-12b-sft-v8-7k-steps --output_dir /home/feil_m/tmp-ct2fast-pythia-12b-sft-v8-7k-steps --force --copy_files tokenizer.json README.md tokenizer_config.json generation_config.json special_tokens_map.json .gitattributes --quantization float16 |
|
|
``` |
|
|
|
|
|
Checkpoint compatible to [ctranslate2>=3.13.0](https://github.com/OpenNMT/CTranslate2) and [hf-hub-ctranslate2>=2.0.6](https://github.com/michaelfeil/hf-hub-ctranslate2) |
|
|
- `compute_type=int8_float16` for `device="cuda"` |
|
|
- `compute_type=int8` for `device="cpu"` |
|
|
|
|
|
```python |
|
|
from hf_hub_ctranslate2 import TranslatorCT2fromHfHub, GeneratorCT2fromHfHub |
|
|
from transformers import AutoTokenizer |
|
|
|
|
|
model_name = "michaelfeil/ct2fast-pythia-12b-sft-v8-7k-steps" |
|
|
# use either TranslatorCT2fromHfHub or GeneratorCT2fromHfHub here, depending on model. |
|
|
model = GeneratorCT2fromHfHub( |
|
|
# load in int8 on CUDA |
|
|
model_name_or_path=model_name, |
|
|
device="cuda", |
|
|
compute_type="int8_float16", |
|
|
tokenizer=AutoTokenizer.from_pretrained("OpenAssistant/pythia-12b-sft-v8-7k-steps") |
|
|
) |
|
|
outputs = model.generate( |
|
|
text=["How do you call a fast Flan-ingo?", "User: How are you doing? Bot:"], |
|
|
) |
|
|
print(outputs) |
|
|
``` |
|
|
|
|
|
# Licence and other remarks: |
|
|
This is just a quantized version. Licence conditions are intended to be idential to original huggingface repo. |
|
|
|
|
|
# Original description |
|
|
|
|
|
- base model: [OpenAssistant/pythia-12b-pre-v8-12.5k-steps](https://huggingface.co/OpenAssistant/pythia-12b-pre-v8-12.5k-steps) |
|
|
- wandb: https://wandb.ai/open-assistant/supervised-finetuning/runs/pcw1ejda |
|
|
- [sampling report](https://raw.githubusercontent.com/Open-Assistant/oasst-model-eval/main/sampling_reports/oasst-sft/2023-05-07_OpenAssistant_pythia-12b-sft-v8-7k-steps_sampling_noprefix2.json) |
|
|
|
|
|
``` |
|
|
pythia-12b-sft-8: |
|
|
dtype: fp16 |
|
|
log_dir: "pythia_log_12b" |
|
|
learning_rate: 6e-6 |
|
|
model_name: OpenAssistant/pythia-12b-pre-v8-12.5k-steps |
|
|
output_dir: pythia_model_12b |
|
|
weight_decay: 0.0 |
|
|
residual_dropout: 0.0 |
|
|
max_length: 2048 |
|
|
use_flash_attention: true |
|
|
warmup_steps: 100 |
|
|
gradient_checkpointing: true |
|
|
gradient_accumulation_steps: 2 |
|
|
per_device_train_batch_size: 4 |
|
|
per_device_eval_batch_size: 4 |
|
|
eval_steps: 251 |
|
|
save_steps: 500 |
|
|
num_train_epochs: 8 |
|
|
save_total_limit: 4 |
|
|
num_train_epochs: 8 |
|
|
save_total_limit: 3 |
|
|
use_custom_sampler: true |
|
|
sort_by_length: false |
|
|
save_strategy: steps |
|
|
datasets: |
|
|
- oasst_export: |
|
|
lang: "bg,ca,cs,da,de,en,es,fr,hr,hu,it,nl,pl,pt,ro,ru,sl,sr,sv,uk" |
|
|
input_file_path: 2023-05-06_OASST_labels.jsonl.gz |
|
|
val_split: 0.05 |
|
|
- vicuna: |
|
|
val_split: 0.05 |
|
|
max_val_set: 800 |
|
|
fraction: 0.4 |
|
|
- dolly15k: |
|
|
val_split: 0.05 |
|
|
max_val_set: 300 |
|
|
- grade_school_math_instructions: |
|
|
val_split: 0.05 |
|
|
- code_alpaca: |
|
|
val_split: 0.05 |
|
|
max_val_set: 250 |
|
|
- red_pajama: |
|
|
fraction: 0.05 |
|
|
max_val_set: 1000 |
|
|
- wizardlm_70k: |
|
|
val_split: 0.05 |
|
|
max_val_set: 500 |
|
|
fraction: 0.4 |
|
|
- poem_instructions: |
|
|
fraction: 0.5 |
|
|
val_split: 0.025 |
|
|
``` |
|
|
|
