Text Generation
Transformers
PyTorch
bloom
text-generation-inference
Inference Endpoints

global_step100000 and global_step200000 seem to be exactly the same model

#3
by CharlieTele - opened

I am checking out global_step100000 and global_step200000 for 7b1 model and it seems that they gives exactly the same evaluation results on my metrics. I redownloaded the model and it give still the same results. I suspect that one of the model is wrongly uploaded, and very probably lobal_step100000 is wrong.

Sign up or log in to comment