Thomas Capelle

icon indicating a website link https://tcapelle.github.io/ icon indicating an email [email protected]

icon company W&B icon location Chambery, France icon location Transformers Transformers Transformers...

Results 169 comments of Thomas Capelle

can't run llama-2-7b-hf even though I'm using use_auth_token

Lol I was using: `meta-llama/Llama-2-7B-hf` instead of `meta-llama/Llama-2-7b-hf`...

Add lm-eval-harness colab

should we merge this?

Fix Profiling Pytorch example

Can I merge this @scottire ?

fix: update pytorch-lightning examples

Do you need help?

fix(colabs): Update docs for Report API

Can I merge this?

Did anyone get good CIFAR10 results?

nop, have been trying all day. MNIST works fine

Did anyone get good CIFAR10 results?

I do here, but with a different codebase: [https://wandb.ai/capecape/train\_sd/reports/How-to-Train-a-Conditional-Diffusion-Model-from-Scratch--VmlldzoyNzIzNTQ1][https_wandb.ai_capecape_train_sd_reports_How-to-Train-a-Conditional-Diffusion-Model-from-Scratch--VmlldzoyNzIzNTQ1] Sent from ProtonMail mobile \-------- Original Message -------- On Nov 5, 2022, 6:30 PM, Michael Albergo < ***@***.***> wrote: > >...

Using config_path with subparsers does not seem to work

Is this going to be fixed? is there are workaround?

[Feature Request] Full-Tuning Example

Already, to train in FP16, you need a ton of memory; the 7B param model will need 14GB of memory in FP16 just to load the weights. The gradients would...

Will there be a docker?

here you go! ```docker FROM pytorch/pytorch:2.2.2-cuda12.1-cudnn8-runtime WORKDIR /workspace/torchtune # RUN git config --global --add safe.directory /workspace/torchtune # add this if you have issues COPY . /workspace/torchtune RUN python -m pip...