INDEX

Explanations

Tower

np_max-act · gemini-2.0-flash

This neuron primarily activates on occurrences of the word “Tower” (and its close variations) in the text.

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

bugs

-0.07

.Red

-0.06

.exp

-0.06

bins

-0.06

_move

-0.06

.when

-0.06

 explains

-0.06

aldo

-0.06

England

-0.06

 English

-0.06

POSITIVE LOGITS

 tower

0.17

 Tower

0.17

Tower

0.12

 towers

0.11

 Towers

0.11

 Torres

0.08

tower

0.08

 postav

0.08

 pillar

0.08

innacle

0.08

Activations Density 0.005%

Tower

This neuron primarily activates on occurrences of the word “Tower” (and its close variations) in the text.

No Comments

No Known Activations