File size: 450 Bytes
dc1c3bd f823e16 0120cf9 f823e16 0120cf9 f823e16 0120cf9 f823e16 0120cf9 f823e16 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
---
license: other
---
WizardLM-7B with 10k+ context using Landmark Attention.
Model generated using Landmark-Attention-QLoRA
https://github.com/eugenepentland/landmark-attention-qlora
A merge of the following models:
https://huggingface.co/TheBloke/wizardLM-7B-HF
https://huggingface.co/eugenepentland/WizardLM-7B-Landmark-Attention-QLoRA
Can be loaded in using oobooga, make sure to have the --trust-remote-code option on for it to function. |