Downloads · 30 days
11
1% of all-time downloads
tony10101105/LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps
LatentQA-Llama-3.1-8B-Instruct-Lora-14000steps is a text generation model from tony10101105. Use it when you need the model to write or continue text. It is set up for peft.
[8th March 2026 Update]: Upload a new model (overriding the original one) without adapter on lmhead. Lora adapter on lmhead invokes a subtle issue on Peft that leads to device mismatch when using multi-gpus.
Downloads · 30 days
11
1% of all-time downloads
All-time downloads
1.8K
Public
Repo size
1.4 GB
Likes
0
Public
Click a slice to open those files.
.safetensors168 MB · 100%
From the Hugging Face model README
[8th March 2026 Update]: Upload a new model (overriding the original one) without adapter on lm_head. Lora adapter on lm_head invokes a subtle issue on Peft that leads to device mismatch when using multi-gpus.
This model replicates LatentQA activation decoder for Llama-3.1-8B-Instruct. The original release of LatentQA decoder is Llama-3-8B-Instruct-based. This model has the same explantion results on the "promote_veganism" example documented in the repo of original work.
Activation decoder to interpret Llama-3.1-8B-Instruct's internal activations with natural language descriptions.
Please refer to https://github.com/aypan17/latentqa
The LatentQA fine-tuning is not extensive. The decoder is only good at answering questions analogous to fine-tuning dataset.