TheSkullery/LD-Zephyria-37b, healed by training a 128 rank LoRA (targeting only the q and down projections) on one epoch of TheSkullery/Aether-Lite-PurHyDe

Thanks to @Steelskull for the merge recipe and dataset.

Downloads last month
7
Safetensors
Model size
37.5B params
Tensor type
BF16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for athirdpath/LD-Zephyria-37b_HEALED

Finetuned
(1)
this model
Finetunes
1 model
Quantizations
2 models

Dataset used to train athirdpath/LD-Zephyria-37b_HEALED