INDEX

Explanations

remove

np_max-act · gemini-2.0-flash

The neuron strongly activates on tokens for removal operations—e.g. “remove,” “Remove,” or “unregister.”

oai_token-act-pair · o4-mini Triggered by @xinyanhu8

New Auto-Interp

Configuration

andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1

Dataset (Dashboard)

Various

Features

131,072

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 discriminator

-0.06

 detection

-0.06

 Utilities

-0.06

 сф

-0.06

 containers

-0.06

 initiation

-0.06

quarters

-0.06

 raids

-0.06

 synthesis

-0.06

 vectors

-0.06

POSITIVE LOGITS

 ApiResponse

0.07

 удал

0.07

enger

0.06

 -->
↵
↵

0.06

esinde

0.06

 seviy

0.06

.FindAsync

0.06

议

0.06

(sorted

0.06

elly

0.06

Activations Density 0.026%

remove

The neuron strongly activates on tokens for removal operations—e.g. “remove,” “Remove,” or “unregister.”

No Comments

No Known Activations

remove

The neuron strongly activates on tokens for removal operations—e.g. “remove,” “Remove,” or “unregister.”

No Comments

No Known Activations