INDEX

Explanations

was

np_max-act · gemini-2.0-flash

This neuron fires on the answer-opening phrases that signal a discovered solution (e.g. “Turns out…” or “I found out…”).

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

演

-0.07

xEA

-0.07

 đài

-0.06

集合

-0.06

-being

-0.06

iado

-0.06

 Barker

-0.06

يمي

-0.06

.LocalDate

-0.06

.preference

-0.06

POSITIVE LOGITS

 helf

0.06

/?

0.06

.savefig

0.06

leg

0.06

σσότε

0.06

antas

0.06

.yy

0.06

/g

0.06

umsuz

0.06

sn

0.06

Activations Density 0.032%

was

This neuron fires on the answer-opening phrases that signal a discovered solution (e.g. “Turns out…” or “I found out…”).

No Comments

No Known Activations