# Probing fine-tuned model

**URL:** <https://discuss.huggingface.co/t/probing-fine-tuned-model/2393>\
**Category:** Beginners\
**Created:** [December 1, 2020, 3:59pm UTC](https://discuss.huggingface.co/t/probing-fine-tuned-model/2393 "2020-12-01T15:59:17Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![thak123](https://avatars.discourse-cdn.com/v4/letter/t/d07c76/32.png) [@thak123](https://discuss.huggingface.co/u/thak123)\
**Post date:** [December 1, 2020, 3:59pm UTC](https://discuss.huggingface.co/t/probing-fine-tuned-model/2393/1 "2020-12-01T15:59:17Z")

</div>

Any tools to see how the model weighs the various words/pieces in the input for the predictions.

---

<div class="post-metadata">

**Author:** ![rgwatwormhill](https://avatars.discourse-cdn.com/v4/letter/r/a9adbd/32.png) [@rgwatwormhill](https://discuss.huggingface.co/u/rgwatwormhill)\
**Post date:** [December 3, 2020, 11:20am UTC](https://discuss.huggingface.co/t/probing-fine-tuned-model/2393/2 "2020-12-03T11:20:59Z")

</div>

hi thak123,

If you are using BERT model in PyTorch, then BertViz by Jesse Vig is quite nice.

> **[GitHub - jessevig/bertviz: BertViz: Visualize Attention in NLP Models (BERT,...](https://github.com/jessevig/bertviz)**
>
> BertViz: Visualize Attention in NLP Models (BERT, GPT2, BART, etc.) - GitHub - jessevig/bertviz: BertViz: Visualize Attention in NLP Models (BERT, GPT2, BART, etc.)