# 403 error on dataset fineweb-2

**URL:** </static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2F403-error-on-dataset-fineweb-2%2F168620%3E%5C
**Category:** 🤗Datasets\
**Created:** [September 23, 2025, 9:45pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2F403-error-on-dataset-fineweb-2%2F168620 "2025-09-23T21:45:26Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![blazux](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/blazux/32/54198_2.png) [@blazux](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fblazux)%5C
**Post date:** [September 23, 2025, 9:45pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2F403-error-on-dataset-fineweb-2%2F168620%2F1 "2025-09-23T21:45:26Z")

</div>

Hi,

I was training a small model just for fun when the error occured (after more 100k steps) :

requests.exceptions.HTTPError: 403 Client Error: Forbidden for url: [https://huggingface.co/datasets/HuggingFaceFW/fineweb-2/resolve/a8a99b128121a41b17d95901715603386f6b1daf/data/fra\_Latn/train/000\_00000.parquet](https://huggingface.co/datasets/HuggingFaceFW/fineweb-2/resolve/a8a99b128121a41b17d95901715603386f6b1daf/data/fra_Latn/train/000_00000.parquet)

I’m wondering if I have reach some rate limits or else ? I guess it shoul failed way earlier if I was doing it wrong ?

I’m using it with streaming on:

```auto
    ds_fr = load_dataset(
        "HuggingFaceFW/fineweb-2",
        name="fra_Latn",
        split="train",
        streaming=True
    )

```

Any idea what the problem can be ?

Thanks,

---

<div class="post-metadata">

**Author:** ![John6666](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/john6666/32/27664_2.png) [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [September 24, 2025, 12:35am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2F403-error-on-dataset-fineweb-2%2F168620%2F2 "2025-09-24T00:35:14Z")

</div>

> HTTPError: 403 Client Error: Forbidden for url

When `streaming=True`, shards are fetched on-demand, so it’s not unusual for errors to occur midway through fetching. Judging from the error message, it appears to be a CDN or network error, so I don’t think it’s a code issue.

Since the retry limit is likely less restrictive during login, how about [doing `huggingface_hub.login()` beforehand during training](https://huggingface.co/docs/huggingface_hub/main/quick-start#authentication) and [configuring `datasets` settings like increasing the retry count](https://github.com/huggingface/datasets/issues/6172) to enhance error tolerance?

Although I don’t think it’s the case this time, it’s not unheard of for the dataset repository to be updated while streaming the dataset—a rare scenario. To avoid this, explicitly specifying the revision would be the surest way.

---

<div class="post-metadata">

**Author:** ![system](https://us1.discourse-cdn.com/hellohellohello/original/2X/d/de4155eb4aa4108ecb32a1389d7cc37ae69f88b7.png) [@system](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fsystem)%5C
**Post date:** [September 27, 2025, 2:06pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2F403-error-on-dataset-fineweb-2%2F168620%2F3 "2025-09-27T14:06:23Z")

</div>

Closed on September 27, 2025, 2:06pm UTC

This topic was automatically closed 12 hours after the last reply. New replies are no longer allowed.