forum
General advice for instruction-tuning a base model
Set a seed
triton library not found!
Mistral Nemo 12B 8-bits inference is not working
Run Unsloth on a different GPU
Notebook tutorial for multi-label classification
Out of Memory on RTX 3060 Laptop GPU with 4-bit Quantization
Solved
Bogus results after finetuning
How can I access unsloth pro version?
Google Colab to HuggingFace Error
Open
Google Collab to Ollama gguf
Model does not stop generating during inference after fine-tunning.
ImportError on Paperspace: libcusparse.so.12: undefined symbol
Solved
Model finetuned generate responses without stopping : llama3
save adapter only vs save merged_16bit
LoRA Adapter warnings in vllm
🍲Agent Chef🥘: LORA MERGE ISSUE
Fine tuning with text dataset!
Deep speed support
I have a 3080 with 12 GB of RAM, and I'm trying to fine-tune "unsloth/Phi-3.5-mini-instruct-bnb-4bit
Open
Using the example chat template for OLlama raises RuntimeError
Open
Unsloth models deployment documentation ??????
Open
Fine-tuning multiple times
Just got No module name triton error
Solved
GGUF Upload - Suuuper large file
Change the code instruction template in unsloth to match the gemma 2 instruction template
Open
Unsloth 4bit model - is it something different?
Open
Has anyone else run into this error with the latest Unsloth, accelerate, and TRL regarding DPO?
Erractic bug in Phi 3.5 Did you mean: 'short_cos_cached'?
Bug in DPO Zephr Unsloth Example notebook
whats up with the llama 3.1 tokenizer?
Open
Quantization Issue
Deployment of finetuned llam3 7B model on aws sagemaker jupiterlab notebooks ?
Finetuning Llama Instruct model - DataCollatorForSeq2Seq error
Finetune Nemo 12b with ShareGPT conversational style dataset
cpu
AttributeError: 'UnslothTrainingArguments' object has no attribute 'packing'
LM studio does not recognize my model
Solved
Does unsloth provide "train_on_inputs"= False parameter similar to axolotl?
Open
how to connect trained model with unsloth to RAG and How to teach unsloth model to say"I don't know"
I want to change the training from 4bit to 16bit
continue fine tuning after outputting the adapter
Production inference
Chatml for Qwen Model
Hi. Just tried to perform fine tuning on tinyllama but only the model unsloth/tinyllama-chat-bnb-4b
Data Preparation Tutorial for Fine-Tuning a Model
Unsloth with TRL DPO
Solved
Issues with back propogatting a loss through a lora adapted unsloth/tinyllama-bnb-4bit model
Data preparation for Chat. I only have long chunks of proprietary text data.
Open
Solved
Hosting model on Huggingface's Inference Endpoints?
Open