Mengwei Liu
Mengwei Liu
> The memory offset of aten.sin should have been 256 instead of 512. I am not sure why we have a hole of 256 bytes between x and aten.sin. The...
`detach` should be removed from the graph, not sure why it sticks. For `alias`, my understanding is functionalization may not replace it with an `alias_copy` if we never change the...
Link to landed trunk PR: * https://github.com/pytorch/executorch/pull/3782 Link to release branch PR: * https://github.com/pytorch/executorch/pull/3806 Criteria category and reasons: Doc improvement based on users feedback ---------- @mergennachin merged
Link to landed trunk PR: * https://github.com/pytorch/executorch/pull/3787 (seems like this is already merged into `release/0.2`?) Link to release branch PR: * https://github.com/pytorch/executorch/pull/3787 Criteria category and reasons: Fix torch and torchaudio...
I think @antmikinka was able to finish exporting, if not please file another issue. Closing.
@l3utterfly Can you share the full error message? I thought it happens at `load_state_dict` but it seems `strict=False` so it shouldn't error out. https://github.com/pytorch/executorch/blob/main/examples/models/llama2/model.py#L197
@l3utterfly vocab_size is something configurable. Can you change the value in `/home/layla/src/text-generation-webui/models/Meta-Llama-3-8B-Instruct/original/params.json` to the new one and retry?
@l3utterfly thanks for offering help! We have been talking about implementing different kv cache manipulation techniques but haven't got a chance to that part. For now you can look at...
This looks good. Can you add some unit test for the verifier?