forum

1495 threads · Page 10 of 30

In the gpt-oss notebook, the finetuned model doesn't seem to use French for reasoning? 10 messages
Errors on latest release 10 messages
Loss function to use for JSON outputs. 8 messages
Open
gpt-oss-20b continued pretraining 50 messages
Unsloth Index Out of Bound 3 messages
Open
Error importing unsloth 3 messages
Open
modules_to_save with vllm 5 messages
Missing formatting_func error 2 messages
Open
AcceleratorError: CUDA error: CUDA-capable device(s) is/are busy or unavailable 3 messages
Solved
Creating Dynamic 2.0 GGUFs for custom models 52 messages
Open
How to save a full fine tuned model 3 messages
Issues with GLM-4.5-AIR 7 messages
Open
Feature Request: Sequential Multi-Turn Finetuning Support 4 messages
Open
Continued pretraining and training loss for Gemma 3 2 messages
Open
any guide for train base to chat llm? 9 messages
Medgemma 4b Vision 38 messages
got ZeroDivisionError with qwen3 train_on_responses_only setup 2 messages
Solved
Unable to DPO training and GGUF saving. 30 messages
Open
Function is missing in dataloader "fetch_video" 6 messages
gpt-oss-20B runtime error 5 messages
Open
I want to provide computing resources 2 messages
Bug when apply_chat_template with unsloth/Mistral-Small-3.2-24B-Instruct-2506 2 messages
No module named 'transformers.models.gpt' 6 messages
AttributeError: 'NoneType' object has no attribute 'shape' 4 messages
Open
Unsloth doesn't detect rtx5060 on Windows 4 messages
Solved
Loading unsloth model after fine-tuning. 2 messages
Does Unsloth use HF Transformers, or llama.cpp for inference 5 messages
Model is not trained when converted to GGUF/SFTrainer recommended settings for Gemma 3 1b (Q&A Data) 45 messages
gpt-oss-120b ollama template 19 messages
llama.cpp optimization help for gpt-oss-120b 2 messages
Suggestion for Mistral Small 3.2 3 messages
GPT-OSS sinks with Flex attn for LC 4 messages
BackendCompilerFailed 20 messages
Solved
Loading a Merged Gemma 3-270b-it model 3 messages
Open
Gemma3 notebook error 14 messages
SFT and GRPO on local RTX 3090 12 messages
Are vision models supported for fast_inference=True? 6 messages
Fine tuning 3 messages
Open
Size mismatch after training unsloth_classification 3 messages
vision model q8_0 3 messages
How to properly retrain a model to a new language with tokenizer replacement? 3 messages
Llama-3.2-11B 8 messages
The size of tensor 2 messages
no logs 3 messages
Add proposed ReadMe for GLM 4.5 (Fixes MLX This Repo) 3 messages
Solved
RuntimeError: Output 0 of UnslothFusedLossBackward is a view 4 messages
Open
synthetic data notebook 2 messages
performance low budget graphic card 8 messages
How to load in a full fine tuned model? 34 messages
Open
Version mismatch on latest libs and packs 9 messages