Quantization made by Richard Erkhov.

gpt2-medium-italian-embeddings - GGUF

Model creator: https://huggingface.co/GroNLP/
Original model: https://huggingface.co/GroNLP/gpt2-medium-italian-embeddings/

Name	Quant method	Size
gpt2-medium-italian-embeddings.Q2_K.gguf	Q2_K	0.14GB
gpt2-medium-italian-embeddings.IQ3_XS.gguf	IQ3_XS	0.16GB
gpt2-medium-italian-embeddings.IQ3_S.gguf	IQ3_S	0.16GB
gpt2-medium-italian-embeddings.Q3_K_S.gguf	Q3_K_S	0.16GB
gpt2-medium-italian-embeddings.IQ3_M.gguf	IQ3_M	0.17GB
gpt2-medium-italian-embeddings.Q3_K.gguf	Q3_K	0.19GB
gpt2-medium-italian-embeddings.Q3_K_M.gguf	Q3_K_M	0.19GB
gpt2-medium-italian-embeddings.Q3_K_L.gguf	Q3_K_L	0.2GB
gpt2-medium-italian-embeddings.IQ4_XS.gguf	IQ4_XS	0.19GB
gpt2-medium-italian-embeddings.Q4_0.gguf	Q4_0	0.2GB
gpt2-medium-italian-embeddings.IQ4_NL.gguf	IQ4_NL	0.2GB
gpt2-medium-italian-embeddings.Q4_K_S.gguf	Q4_K_S	0.2GB
gpt2-medium-italian-embeddings.Q4_K.gguf	Q4_K	0.22GB
gpt2-medium-italian-embeddings.Q4_K_M.gguf	Q4_K_M	0.22GB
gpt2-medium-italian-embeddings.Q4_1.gguf	Q4_1	0.22GB
gpt2-medium-italian-embeddings.Q5_0.gguf	Q5_0	0.24GB
gpt2-medium-italian-embeddings.Q5_K_S.gguf	Q5_K_S	0.24GB
gpt2-medium-italian-embeddings.Q5_K.gguf	Q5_K	0.26GB
gpt2-medium-italian-embeddings.Q5_K_M.gguf	Q5_K_M	0.26GB
gpt2-medium-italian-embeddings.Q5_1.gguf	Q5_1	0.26GB
gpt2-medium-italian-embeddings.Q6_K.gguf	Q6_K	0.28GB

Original model description:

language: it tags:

adaption
recycled
gpt2-medium pipeline_tag: text-generation

GPT-2 recycled for Italian (medium, adapted lexical embeddings)

Wietse de Vries • Malvina Nissim

Model description

This model is based on the medium OpenAI GPT-2 (gpt2-medium) model.

The Transformer layer weights in this model are identical to the original English, model but the lexical layer has been retrained for an Italian vocabulary.

For details, check out our paper on arXiv and the code on Github.

Related models

Dutch

gpt2-small-dutch-embeddings: Small model size with only retrained lexical embeddings.
gpt2-small-dutch: Small model size with retrained lexical embeddings and additional fine-tuning of the full model. (Recommended)
gpt2-medium-dutch-embeddings: Medium model size with only retrained lexical embeddings.

Italian

gpt2-small-italian-embeddings: Small model size with only retrained lexical embeddings.
gpt2-small-italian: Small model size with retrained lexical embeddings and additional fine-tuning of the full model. (Recommended)
gpt2-medium-italian-embeddings: Medium model size with only retrained lexical embeddings.

How to use

from transformers import pipeline

pipe = pipeline("text-generation", model="GroNLP/gpt2-medium-italian-embeddings")

from transformers import AutoTokenizer, AutoModel, TFAutoModel

tokenizer = AutoTokenizer.from_pretrained("GroNLP/gpt2-medium-italian-embeddings")
model = AutoModel.from_pretrained("GroNLP/gpt2-medium-italian-embeddings")  # PyTorch
model = TFAutoModel.from_pretrained("GroNLP/gpt2-medium-italian-embeddings")  # Tensorflow

BibTeX entry

@misc{devries2020good,
      title={As good as new. How to successfully recycle English GPT-2 to make models for other languages}, 
      author={Wietse de Vries and Malvina Nissim},
      year={2020},
      eprint={2012.05628},
      archivePrefix={arXiv},
      primaryClass={cs.CL}
}