Huggingface bertscore metric GPU leak?

I'm using bertscore as a metric for my project, and I'm getting a GPU leak that I can't understand. I've been able to reproduce the issue with this simple script, looking at GPU usage on my GPU.

Evaluate version is 0.2.2, python=3.8.

```python
import evaluate

predictions = ['a' for _ in range(30)]
references = ['b' for _ in range(30)]

# nvidia-smi output
# Sat Apr 29 12:25:16 2023
# +-----------------------------------------------------------------------------+
# | NVIDIA-SMI 510.108.03 Driver Version: 510.108.03 CUDA Version: 11.6 |
# |-------------------------------+----------------------+----------------------+
# | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
# | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
# | | | MIG M. |
# |===============================+======================+======================|
# | 0 NVIDIA GeForce ... Off | 00000000:07:00.0 Off | N/A |
# | N/A 41C P8 2W / N/A | 1356MiB / 8192MiB | 1% Default |
# | | | N/A |
# +-------------------------------+----------------------+----------------------+
# ...
# nothing for python

bertscore = evaluate.load("bertscore")
scores = bertscore.compute(predictions=predictions, references=references, lang='en', batch_size=64)['f1']

# nvidia-smi output:
# +-----------------------------------------------------------------------------+
# | NVIDIA-SMI 510.108.03 Driver Version: 510.108.03 CUDA Version: 11.6 |
# |-------------------------------+----------------------+----------------------+
# | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
# | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
# | | | MIG M. |
# |===============================+======================+======================|
# | 0 NVIDIA GeForce ... Off | 00000000:07:00.0 Off | N/A |
# | N/A 41C P8 2W / N/A | 3106MiB / 8192MiB | 1% Default |
# | | | N/A |
# +-------------------------------+----------------------+----------------------+
# ...
# | 0 N/A N/A 1087985 C python 1751MiB |
# +-----------------------------------------------------------------------------+

del scores
del bertscore

# nvidia-smi output:
# +-----------------------------------------------------------------------------+
# | NVIDIA-SMI 510.108.03 Driver Version: 510.108.03 CUDA Version: 11.6 |
# |-------------------------------+----------------------+----------------------+
# | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
# | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
# | | | MIG M. |
# |===============================+======================+======================|
# | 0 NVIDIA GeForce ... Off | 00000000:07:00.0 Off | N/A |
# | N/A 41C P8 2W / N/A | 3106MiB / 8192MiB | 1% Default |
# | | | N/A |
# +-------------------------------+----------------------+----------------------+
# ...
# | 0 N/A N/A 1087985 C python 1751MiB |
# +-----------------------------------------------------------------------------+
```

Does anyone know what's causing this issue? I'm expecting the GPU utilisation to go down after deleting both the scores and the model.