MaziyarPanahi/Llama-3-8B-Instruct-DPO-v0.3-32k-GGUF · Discussions

Resources

View closed (2)

Warning: llm_load_vocab: missing pre-tokenizer type, using: 'default'

#4 opened 7 months ago by

The f16 with 32k ctx fits nicely in 24GB VRAM

#3 opened 8 months ago by