minGPT-tuned
minGPT-tuned copied to clipboard
A *tuned* minimal PyTorch re-implementation of the OpenAI GPT (Generative Pretrained Transformer) training
self.time_weight = nn.Parameter(torch.ones(self.n_heads, self.block_length, self.block_length)) what is the block_size?
Thank-you for these enhancements. I'm running the play_char notebook, but keep getting errors on the first epoch completion about test_loss not being defined before it is assigned. I tried to...
Hi, Thanks for making your enhancements to minGPT available. I am curious why your play_math model will deal with ndigit = 4 when anything above ndigit = 3 with stock...