Update README.md
Browse filesThis is a 30 million parameter model trained on approximately 2 billion tokens of web data from the uncleaned dataset for 1 epoch using my custom torch dataset class. This model has a context length of 512.
This is a 30 million parameter model trained on approximately 2 billion tokens of web data from the uncleaned dataset for 1 epoch using my custom torch dataset class. This model has a context length of 512.