--- license: apache-2.0 datasets: - oscar-corpus/OSCAR-2109 language: - en - nl pipeline_tag: text-generation library_name: transformers --- # B-GPT_en_nl_sequential The B-GPT Models are bilingual GPT-2 style models. For the first half of training, this model was trained only on English data. In the second half of training, the model was trained on only {language_2} data.. At the end of training, 50 % of training data seen by the model is English and 50 % is Dutch. The tokenizer was trained on the same proportions of English and Dutch data. ## Model details: All models are trained with a [CLS] (same as [BOS]) token prepended, and a [SEP] (same as [EOS]) token separating sequences. For best results, make sure that [CLS] is prepended to your input sequence (see sample usage linked above)! Details for this model specifically: * Architecture: gpt2 * Parameters: 124770816 * Maximum sequence length: 512 tokens * Training text data (raw): [XXXX] * Training tokens: 12B * Vocabulary size: 50000 * Compute cost: ~9 NVIDIA A6000 GPU hours * CO2 Emission: 1.17 kg Training datasets (percentages prior to deduplication): * 100.00000%: [OSCAR 2021/09](https://huggingface.co/datasets/oscar-corpus/OSCAR-2109) Checkpoints are taken at training steps: 0, 10000, 20000, 30000, 40000, 50000, 64000, 64010, 64020, 64030, 64040, 64050, 64060, 64070, 64080, 64090, 64100, 64110, 64120, 64130, 64140, 64150, 64160, 64170, 64180, 64190, 64200, 64300, 64400, 64500, 64600, 64700, 64800, 64900, 65000, 66000, 67000, 68000, 69000, 70000, 80000, 90000, 100000, 110000, 120000, 128000. ## Use This Model Load the model: ``` from transformers import AutoTokenizer, AutoModel tokenizer = AutoTokenizer.from_pretrained("B-GPT_en_nl_sequential") model = AutoModel.from_pretrained("B-GPT_en_nl_sequential") ```` Text Generation: ``` from transformers import pipeline pipe = pipeline("text-generation", model="B-GPT_en_nl_sequential") pipe("I am a") ``` ## Citation If you use this model, please cite: ``` ```