Updating README with clean_up_tokenization_spaces

README.md

@@ -11,14 +11,14 @@ datasets:
 - HPLT/HPLT2.0_cleaned
 ---
 
-# HPLT Bert for Spanish
+# HPLT v2.0 BERT for Spanish
 
 <img src="https://hplt-project.org/_next/static/media/logo-hplt.d5e16ca5.svg" width=12.5%>
 
 This is one of the encoder-only monolingual language models trained as a second release by the [HPLT project](https://hplt-project.org/).
 It is a so called masked language model. In particular, we used the modification of the classic BERT model named [LTG-BERT](https://aclanthology.org/2023.findings-eacl.146/).
 
-A monolingual LTG-BERT model is trained for some languages in the [HPLT 2.0 data release](https://hplt-project.org/datasets/v2.0).
+We present monolingual LTG-BERT models for more than 50 languages out of 191 total in the [HPLT v2.0 dataset](https://hplt-project.org/datasets/v2.0).
 
 All the HPLT encoder-only models use the same hyper-parameters, roughly following the BERT-base setup:
 - hidden size: 768
@@ -32,7 +32,7 @@ Every model uses its own tokenizer trained on language-specific HPLT data.
 
 [The training statistics of all runs](https://api.wandb.ai/links/ltg/kduj7mjn)
 
-## Example usage
+## Example usage (tested with `transformers==4.46.1` and `tokenizers==0.20.1`)
 
 This model currently needs a custom wrapper from `modeling_ltgbert.py`, you should therefore load the model with `trust_remote_code=True`.
 
@@ -49,7 +49,7 @@ output_p = model(**input_text)
 output_text = torch.where(input_text.input_ids == mask_id, output_p.logits.argmax(-1), input_text.input_ids)
 
 # should output: '[CLS] It's a beautiful place.[SEP]'
-print(tokenizer.decode(output_text[0].tolist()))
+print(tokenizer.decode(output_text[0].tolist(), clean_up_tokenization_spaces=True))
 ```
 
 The following classes are currently implemented: `AutoModel`, `AutoModelMaskedLM`, `AutoModelForSequenceClassification`, `AutoModelForTokenClassification`, `AutoModelForQuestionAnswering` and `AutoModeltForMultipleChoice`.