gpt2-large-dutch / README.md
yhavinga's picture
Saving weights and log at step 1080000
d40a1b1
|
raw
history blame
1.5 kB
metadata
language: nl
widget:
  - text: In het jaar 2030 zullen we
  - text: Toen ik gisteren volledig in de ban was van
  - text: >-
      Studenten en leraren van de Bogazici Universiteit in de Turkse stad
      Istanbul
  - text: In Israël was een strenge lockdown
tags:
  - gpt2-large
  - gpt2
pipeline_tag: text-generation
datasets:
  - yhavinga/mc4_nl_cleaned

GPT2-Large pre-trained on cleaned Dutch mC4 🇳🇱

Dataset:

Tokenizer:

  • Tokenizer trained on mC4 with scripts from the Huggingface Transformers Flax examples

Training details:

  • Training started on step 360K (bs 16) ppl 21 of earlier model trained with Adam optimizer.
  • Training at step 1080K of 2082K (52%) ppl 15,1
  • Block size: 512
  • Optimizer: adafactor
  • Learning rate: 3.3e-5
  • Batch size: 32
  • Warmup steps: 5000

Work in progress. Dec 2021-Jan2022