Hello everyone,
I have trained a transformer model as a sequence-to-sequence model, did not finetune an LLM.
However, my way of saving the model into huggingface or even locally seems to be incorrect, meaning that checkpoints, weights and other important model configurations may have not been saved properly.
because when I try to load it from huggingface, it gives me this error:
ValueError: Unrecognized model in Lu… Should have a model_type key in its config.json, or contain one of the following strings in its name: albert, align, altclip…
Please find the layout of my work in this notebook: