lxe / simple-llm-finetuner

Simple UI for LLM Model Finetuning

lxe/simple-llm-finetuner Issues

How should I prepare the dataset for generative question answering on the private documents?
Updated 3 months ago47
will this work with quantized GGUF files?
Updated 7 months ago
[Request] Retrain adapter from checkpoint?
Updated 7 months ago
Performance after FineTuning
Updated 9 months ago3
About llama-2-70B fine-tuning
Updated 9 months ago
Getting the repo id error from the web interface
Updated 9 months ago
How to use CPU instead of GPU
Updated 10 months ago2
AMD GPU compability or CPU
Updated 10 months ago
RuntimeError: expected scalar type Half but found Float
Updated 10 months ago3
M1/M2 Metal support?
Updated 10 months ago
[Request] Mac ARM support
Updated a year ago
RuntimeError: unscale_() has already been called on this optimizer since the last update().
Closed a year ago3
[Request] QLoRA support
Updated a year ago
In trainer.py, ignore the last token is not suitable for all situations.
Updated a year ago
`LLaMATokenizer` vs `LlamaTokenizer` class names
Updated a year ago5
Question: Is fine tuning suitable for factual answers from custom data, or is it better to use vector databases and use only the relevant chunk in the prompt for factual answers?
Closed a year ago2
Multi GPU running
Updated a year ago
Issue in train in colab
Updated a year ago7
Getting OOM
Updated a year ago2
How do I merge trained Lora an Llama7b weight?
Updated a year ago2
Error: Adapter lora/decapoda-research_llama-{ADAPTER_NAME} not found.
Closed a year ago
Suggestion to improve UX
Updated a year ago3
"The tokenizer class you load from this checkpoint is 'LLaMATokenizer'."
Closed a year ago3
Verbose function to find out what leads to crash during training?
Closed a year ago1
How can I use the finetuned model with text-generation-webui or KoboldAI?
Updated a year ago4
question: could the model trained be used for alpaca.cpp?
Updated a year ago2
Slow generation speed: around 10 minutes / loading forever on rtx3090 with 64gb ram....
Updated a year ago3
AttributeError: type object 'Dataset' has no attribute 'from_list'
Updated a year ago3
"error" in training - AttributeError: 'CastOutputToFloat' object has no attribute 'weight', RuntimeError: Only Tensors of floating point and complex dtype can require gradients
Updated a year ago5
Training using long stories instead of question/response
Updated a year ago3
how to finetune with 'system information'
Updated a year ago1
Attempting to use 13B in the simple tuner -
Updated a year ago2
Not a problem - but like people should know
Updated a year ago2
Error during Training RuntimeError: mat1 and mat2 shapes cannot be multiplied (511x2 and 3x4096)
Updated a year ago2
Traceback during inference.
Updated a year ago8
Examples to get started with
Updated a year ago4
How the finetuning output looks like?
Closed a year ago1
Question: Native windows support
Updated a year ago2
Can Nivdia 3090 with 24G video memory support finetune?
Updated a year ago4
Inference doesn't work after training
Closed a year ago2
Finetuning in unsupported language
Updated a year ago2
Host on Hugging Face Spaces
Closed a year ago4
Inference output text keeps running on...
Updated a year ago1
Inference works just once
Closed a year ago12
(WSL2) - No GPU / Cuda detected....
Closed a year ago6
Collecting info on memory requirements
Updated a year ago1
Is CUDA 12.0 supported?
Updated a year ago1
Where are the downloaded ".bin" files for the llama model stored on the disk?
Closed a year ago2