h2oai / h2o-llmstudio

H2O LLM Studio - a framework and no-code GUI for fine-tuning LLMs. Documentation: https://docs.h2o.ai/h2o-llmstudio/
https://h2o.ai
Apache License 2.0
4.03k stars 418 forks source link

[DOC] Add to FAQ, facing OSError: Consistency check failed in experiment #860

Closed pascal-pfeiffer closed 1 month ago

pascal-pfeiffer commented 2 months ago

📃 Documentation

If facing an error like this in the training

│ OSError: Consistency check failed: file should be of size 4999819336 but has size 14099570832 ((…)ve/main/model-00002-of-00003.safetensors). 

The most likely reason is network issues either on user side or on Hugging Face Hub side. In most cases, rerunning the experiment will solve it, as the download of the model weights will be re initiated.

Let's add this to the FAQ

sherenem commented 1 month ago

Ack. If we have more error related advice like this, I would love to add a "Troubleshooting guide" to the LLM Studio docs.