forum

1495 threads · Page 26 of 30

Is there any way to split one model across two gpus whilst training? or plans to do so? 4 messages
TypeError: is_bf16_supported() got an unexpected keyword argument 'including_emulation' 4 messages
Solved
Deploying using VLLM 3 messages
Problem in running on google colab T4 4 messages
Hi all, 55 messages
unsloth/Meta-Llama-3.1-405B-bnb-4bit is broken 2 messages
Is there any colab notebook for inferencing unsloth model using vllm 32 messages
Solved
Hallucination issue 28 messages
Open
undefinedError: 'dict object' has no attribute 'value' 24 messages
Unable to run unsloth finetuned model with ollama 3 messages
Open
invalid model reference 4 messages
Running into LLVM ERROR: Cannot select: intrinsic %llvm.nvvm.shfl.sync.bfly.i32 2 messages
Open
Reporting a bug 4 messages
Open
Why Top_P set to 1 is more deterministic than Top_P set to 0.9? 18 messages
Tokenizer 15 messages
Open
The fastest LLM inference on the server 16 messages
Open
Can I retrain my model? 10 messages
Open
AttributeError: torch._dynamo.config.vocab_size does not exist 8 messages
Open
The SFTTrainer parameter `max_steps` can't be `None` 2 messages
gemma-2-2b gives cache size errors 4 messages
Open
Loading model on 2 GPUs for inference 5 messages
Open
Huggingface Inference Endpoints - Chat Template error 7 messages
Open
Error with Model Loading in Ollama 3 messages
Unknown RoPE scaling type dynamic (internlm2.5) 8 messages
Reward training for chat 3 messages
Meta Tensor error while training 5 messages
Finetune on completions only and export to Ollama? 27 messages
Open
TypeError: LlamaRotaryEmbedding.__init__() got an unexpected keyword argument 'config' 3 messages
Not an error, but Unsloth cannot patch MLP layers with our manual autograd engine since either LoRA 11 messages
What does `max_seq_length` do? Is it related to the context window? 87 messages
VRAM Explosion During Loop Training - Help Needed! 51 messages
Solved
Strange output of Meta-Llama-3.1-8B-bnb-4bit after fine tuning 19 messages
Unsloth: `unsloth/Meta-Llama-3.1-8B-bnb-4bit` is not a base model or a PEFT model. 9 messages
llama3.1 - Continuous Pretraining - Rope scaling error when loading model. 2 messages
Can Mike/Daniel share how you guys split/mistral-fied the Phi-3 mini model? 6 messages
Try free version on multi-GPU machine . 2 messages
error during finetune for classification 10 messages
hyperparameter choice 7 messages
Fine-tuning for summarization 3 messages
standardize_sharegpt not working 2 messages
train on specific gpu 2 messages
LLM unable to retain knowledge through training. 19 messages
Open
Error creating GGUF quants for Gemma 2 49 messages
Remote Code Error when Loading Model 13 messages
Issue with 16bit training 2 messages
Script for bnb conversion 21 messages
Chat Data Finetuning! 330 messages
Can't save due to collab storage? 12 messages
New PHI-3 mini (june) support request 2 messages
Batch Generation in Unsloth 4 messages