forum

1495 threads · Page 23 of 30

issue in inferencing using streamlit 4 messages
Open
How to load safetensor model weights after saving to hugging face? 2 messages
Open
save_pretrained_merged ruins my model 5 messages
Open
Dora seems to be using twice the memory? 6 messages
Solved
chat template or instruction? 14 messages
llama 3.2v 11B Instruct fine tuning 6 messages
Issue with Inferencing fine tuned Mistral Nemo Instruct model 34 messages
Open
Building memory for chatbot 2 messages
Open
I want to re-use LoRA weight for continuing training. 11 messages
Model stuck in a repeat loop when asking for longer responses 27 messages
Open
Memory usage is higher on WSL? Any idea why? 3 messages
Open
Suffering from massive drops in model accuracy after conversion to .gguf 30 messages
Optimal long-context finetuning 4 messages
Solved
mistral endless generation 3 messages
TypeError: patch_torch_compile() got an unexpected keyword argument 'ignore_errors' 4 messages
Solved
Error in fp16 3 messages
Install Issues 2 messages
Installing in windows, currently impossible ? 31 messages
Solved
Am I preparing my dataset correctly? 2 messages
multi fine tunning 5 messages
train_on_responses_only setting all tokens to -100 ( instruction_part / response_part set wrong?) 7 messages
Unsloth insanely slow 37 messages
Open
Make strict dataset checking optional 3 messages
Open
ModuleNotFoundError: No module named 'unsloth' on T4 5 messages
Solved
Tensor shape error after Llama3.1 8B Instruct template fine-tuning 14 messages
can't load my data dataset to train on it 2 messages
Open
Memorization problem 92 messages
'LlamaForCausalLM' object has no attribute 'max_seq_length' 21 messages
Solved
Problems with fine tune with RAG 7 messages
ValueError: The model did not return a loss from the inputs, 4 messages
Natural text to Sql 4 messages
Newline after EOS token from tokenizer.apply_chat_template 8 messages
Solved
Crash while train half steps 3 messages
Solved
Quantized models of llama 3.2 vision instruct 2 messages
Unsloth: Quantization failed! for (q5_k_m, q4_k_m) 2 messages
New system message not applicated with ollama after finetuning 2 messages
from unsloth import FastLanguageModel 4 messages
Inference UI notebook doesn't work 5 messages
Open
Version of transformers and 'LlamaTokenizerFast' object has no attribute '_ollama_modelfile' 13 messages
to_sharegpt() 11 messages
Dataset without hugging face? 2 messages
Someone help me create a valid dataset 9 messages
Weird answer from llama3.1 model 7 messages
Open
Solved :ModuleNotFoundError: No module named 'huggingface_hub.utils._token' 3 messages
Fine-tuning question | speed of unsloth trainer 2 messages
Finetune Locally 3 messages
Open
Loss calculation on eval_dataset to avoid overfitting. 2 messages
Training works, Validation Fails OOM (With Reproduction Notebook) 3 messages
Open
Model answers differently after importing it into Ollama 30 messages
Open
A100 40GB runs OOM seemingly very easily 31 messages