INDEX

Explanations

forum discussions

np_max-act · gemini-2.0-flash

The neuron activates on the opening words of the assistant’s refusal explanation—specifically the phrase “Creating a narrative that.”

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

 TRAIN

-0.07

듯

-0.07

 клад

-0.06

ワー

-0.06

_income

-0.06

ustomer

-0.06

 histogram

-0.06

tatus

-0.06

.hours

-0.06

垂

-0.06

POSITIVE LOGITS

 creative

0.07

 الجديد

0.07

 raising

0.06

 nombre

0.06

 constructive

0.06

 Epic

0.06

ध

0.06

 outright

0.06

>Nama

0.06

 ทำให

0.06

Activations Density 0.007%

forum discussions

The neuron activates on the opening words of the assistant’s refusal explanation—specifically the phrase “Creating a narrative that.”

No Comments

No Known Activations