Token indices sequence length is longer than the specified maximum sequence length for this model (12065 > 12000). Running this sequence through the model will result in indexing errors

[saitej123]

Token indices sequence length is longer than the specified maximum sequence length for this model (12065 > 12000). Running this sequence through the model will result in indexing errors

Occured train with LLAMA 3 training :

https://colab.research.google.com/drive/1XbH1vpzrGgx75EpDO-mFxXn4G2aatxr-?usp=sharing

[danielhanchen]

@saitej123 I think you fit a OOM out of memory error - the max seq length issue is fine - we auto truncate it in Unsloth