Sweaterdog
Sweaterdog
When I am trying to fine tune large models in Kaggle, which are most models over 8B parameters in 4-bit precision, I get the following error: ``` ValueError: Some modules...
When I was attempting to add specific tokens to the tokenizer for my fine tuned use case, I got an error stating that the matrix was unable to be resized....
**Description** When I run the following code: ```python if True: model.push_to_hub_gguf( "my/Repo", tokenizer, quantization_method = ["q4_k_m", "q8_0", "q5_k_m",], token = "hf_myTokenIputHere", ) ``` 1. **Environment Setup:** - Google Colab -...
# Full title: Make andy 4 default ollama model # Why? Andy-4 is an open-source model designed from the ground up to play Minecraft, it is far superior than any...
Adds STT capabilties using groqcloud's whisper model as well as audio level detection.
Full title: `[DEV ISSUE] Minecollab Instructions not specific, and do not work without a clear reason.` CPU: i9 10900KF GPU: RTX 3070 OS: Linux Mint 22.1 / Windows 11 WSL...
# The premise TPUs are far more efficient than GPUs, especially for AI workloads, and can have significantly more access to high bandwidth memory. This would be immensely beneficial due...
1. Did you update? `I am on the latest version of the docker container` 2. `Colab` or `Kaggle` or local / cloud `local` 3. Number GPUs used, use `nvidia-smi` `1x...
1. Did you update? `Yes` 2. `Colab` or `Kaggle` or local / cloud `Local` 3. Number GPUs used, use `nvidia-smi` `1x RTX 3070` 4. Which notebook? Please link! `N/A` 5....