When will the model be available?
#1
by
pranay-shah
- opened
Hey, have gone through the training scripts in your Github repo. Thanks so much for that. Wondering when would the model be available here? Also, any plans to upload the gemma-infini models?
I've trained several Models but 1M seq len was accidentally missing(due to the ckpt period was too large), so it would take some time for release :(
I think it would take more than >2 weeks to release checkpoint.
Here's the training graph, showing that model is converges. (0.5 epoch was ~ 0.5B tokens)
plz wait and stay tuned! thanks for you attention :)
