# Not able to train the model on auto train
**URL:**
**Author:**  [@yagyanshmittal](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fyagyanshmittal)%5C
**Post date:** [June 6, 2025, 4:53am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fnot-able-to-train-the-model-on-auto-train%2F158202%2F1 "2025-06-06T04:53:21Z")
When I upload a sample file to train the model , I get this notification : “AutoTrain is a paid offering and you will be charged for this action. You can ignore this message if you are running AutoTrain on a local hardware. Are you sure you want to continue?”
When I click on " yes ", It shows error 500.
any idea what needs to be done here ?
---
**Author:**  [@John6666](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FJohn6666)%5C
**Post date:** [June 6, 2025, 11:41am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fnot-able-to-train-the-model-on-auto-train%2F158202%2F2 "2025-06-06T11:41:37Z")
> [@HF auto train dataset load\_metric error](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fhf-auto-train-dataset-load-metric-error%2F157479)%3A
>
> Hello, I am trying to fine-tune the google-bert/bert-base-uncased model with the lhoestq/squad dataset as shown in the documentation for an example of Extraction Question and Answering which is shown in the link below: I have set up my autotrain UI exactly as shown in the screen shot in the above link. I have checked that I have the correct column names. I have tried running this with a CPU and the small T4. However, in both cases I get the following errors: Device 0: Tesla T4 - 2.88MiB/153…
Some people are having trouble with AutoTrain, so why don’t we join them? However, that 500 error might be a bug…
>
>
> I'm trying to train a model using AutoTrain Advanced.
> All steps are completed …correctly, but I consistently get a 500 Internal Server Error when clicking on "Start Training".
>
> My setup
>
> • HuggingFace username: lester-tavelet
>
> !\[Image\](https://github.com/user-attachments/assets/498c9403-d6db-427d-8e53-3abcab366fc1)
>
> 715
> • Project name: My Astro Mirror
> • Model: Mistral-7B-Instruct-v0.1
> • Dataset format: Instruction format (jsonl) — verified and working
> • Training settings: batch size 2, learning rate 2e-5, LoRA enabled
>
> What I've already done:
>
> • Successfully uploaded dataset via AutoTrain UI
> • Verified dataset format (each sample has "instruction", "input", "output")
> • Bound a valid payment method (credit card) — shows active
> • Tried different browsers, network, and several training configurations
>
> The bug
>
> When clicking “Start Training”, the following happens:
>
> • The status says “Build successful”
> • Then it stalls with GET /ui/is\_model\_training HTTP/1.1 200 OK
> • Then the UI shows: 500 Internal Server Error
>
> Here are some screenshots of the behavior:
>
> !\[Error1\](upload\_your\_screenshot\_here)
> !\[Error2\](upload\_your\_second\_screenshot\_here)
>
> !\[Image\](https://github.com/user-attachments/assets/52c59e34-e266-4858-9bc4-5afa5652a616)
>
> What I need
>
> I’d like to know if:
> • My payment status is correctly linked to training access
> • This project has permission to launch Mistral-7B
> • There's any known server-side bug blocking training from starting
>
> Please help me unlock the training. I've prepared all datasets and settings precisely.
>
> Thank you.