-
Notifications
You must be signed in to change notification settings - Fork 5
All issues
Issue creation is restricted in this repository
Issues
is:issue state:open
is:issue state:open
Search results
Training diverges without flash attention
bugSomething isn't workingSomething isn't workingStatus: Open.#47 In simple-stories/simple_stories_train;Models are not trained on n_ctx-length sequences
bugSomething isn't workingSomething isn't workingStatus: Open.#46 In simple-stories/simple_stories_train;Save the accompanying tokenizer after training and verify the push to HF script
enhancementNew feature or requestNew feature or requestStatus: Open.#42 In simple-stories/simple_stories_train;Add from_pretrained() methods to each model
enhancementNew feature or requestNew feature or requestStatus: Open.#38 In simple-stories/simple_stories_train;Support GPT2 architecture in training script
enhancementNew feature or requestNew feature or requestgood first issueGood for newcomersGood for newcomersStatus: Open.#19 In simple-stories/simple_stories_train;Allow for training on custom and tiktoken tokenizer
enhancementNew feature or requestNew feature or requestStatus: Open.#8 In simple-stories/simple_stories_train;