forum

1495 threads · Page 30 of 30

Solving LLVM ERROR: Cannot select: intrinsic %llvm.nvvm.shfl.sync.bfly.i32 2 messages
Solved
TypeError: not a string // when loading tokenizer. 48 messages
Solved
I got `Some modules are dispatched on the CPU or the disk` error 5 messages
Solved
Got error after run `from unsloth import FastLanguageModel` 8 messages
Solved
Merging a LoRa checkpoint 4 messages
Solved
Fine-tuning Mistral-7b-v0.2 on long sequences (2-5k words) 22 messages
Solved
Beginner looking for help 15 messages
Solved
I need LLM Fine tune Help 16 messages
Solved
Sustained memory spike during/after evaluation 6 messages
Solved
Trying to map liyucheng/ShareGPT90K to get_chat_template 14 messages
Solved
shape '[8, 14336]' is invalid for input of size 173981696 27 messages
Solved
Deepspeed-Fastgen 3 messages
Open
Serve over internet with / without gui ? 12 messages
Solved
RuntimeError: element 0 of tensors does not require grad and does not have a grad_fn 61 messages
Solved
16gb GPU OOM with Mistral 7b with custom eval metric 27 messages
Solved
Mistral chat_templates - tokenizer = get_chat_template error 10 messages
Solved
expected string or bytles like object kaggle mistral-7b-v0.2 27 messages
Open
Metadata for project name unknown 7 messages
Solved
Quantisation doesn't work locally. 228 messages
Solved
model works fine, notebook version of q8 works fine, files sent to HF are trash 51 messages
Solved
llama.cpp error 10 messages
Solved
Unsloth Chat Template Error 32 messages
Solved
AttributeError: 'LlamaTokenizerFast' object has no attribute 'original_push_to_hub' 93 messages
Solved
How to deploy Unsloth train model from Huggingface repo? 54 messages
Solved
Uploading Lora adapter to Huggingface fail 8 messages
Solved
Correct format to make finetune on TinyLLama or Mixtral models? 3 messages
Solved
NaN eval loss with mistral notebook 23 messages
Solved
NotImplementedError when running trainer.train() 18 messages
Solved
Further fine tuning a fine tuned model 92 messages
Solved
'LlamaTokenizer' object has no attribute '_tokenizer 11 messages
Open
PyTorch Backward Compatibility 19 messages
Solved
ImportError: cannot import name '_functionalize_sync' from 'torch._utils' 22 messages
Solved
are the gguf quants the old ones or the new imatrix quants? 4 messages
Solved
quant issue 46 messages
Solved
After DPO training with LoRA adapter, generate completions with original model 3 messages
Solved
Attempt unsloth library on Mamba-2.8b 19 messages
Solved
Lowest amount of VRAM needed for finetuning and saving 7b models like Mistral and LLAMA? 15 messages
Solved
cuda errors in kaggle? what just happened 13 messages
Solved
Tiny Llama notebook error at `from unsloth import FastLanguageModel` 5 messages
Solved
Using A100 for gemma training 4 messages
Solved
How to have an AI specialized in my codebase 9 messages
Solved
Load LORA Adapters and use for inference. 7 messages
Solved
Empty/Newline output from the model 14 messages
Solved
First forum error WSL conda install from guide 57 messages
Solved
✨ New Support Forum 31 messages