OpenChatKit
OpenChatKit copied to clipboard
togethercomputer
mamba env create -f environment.yml pytorch/linux-64 Using cache pytorch/noarch Using cache nvidia/linux-64 Using cache nvidia/noarch Using cache conda-forge/linux-64 Using cache conda-forge/noarch Using cache pkgs/main/linux-64 No change pkgs/main/noarch No change pkgs/r/linux-64...
Minor fix for error 'object has no attribute "use_parallel_residual"' that I got
…ble_input_require_grads' *Error:* ```bash (trainer) kristijan@aiweiss-1:~/trainer/OpenChatKit$ python training/lora/redpajama-incite-chat-3b.py ===================================BUG REPORT=================================== Welcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues ================================================================================ CUDA SETUP: CUDA runtime path found: /home/kristijan/.micromamba/envs/trainer/lib/libcudart.so...
This change consolidates the prepare scripts to share a common implementation, with a specific implementations for different models.
- Creates a starting README for LORA finetuning - Adds a general finetune.py that can be used with any model with optional 8-bit and DeeperSpeed support - Adds example notebook...
this could help with CUDA OOM errors especially on consumer grade hardware. prompt and output tensors will be erased from VRAM
**Describe the bug** After following the red panda fine tuning tutorial, running the bot inference script with the output model results in an error. ``` $python ./inference/bot.py --model=model_ckpts/hf/ Loading model_ckpts/hf/...
Sorry, this is not a bug, but I wanted some feedback. I've been able to get the Pythia-Chat-Base-7B model running on very low spec hardware (4G gpu) and it works...
While trying to implement **Pythia-Chat-Base-7B** I am getting this error on running the very fist command (`python inference/bot.py --model togethercomputer/Pythia-Chat-Base-7B`) after creating and activating the conda env. Can anyone help...
