INDEX

Explanations

col

np_max-act · gemini-2.0-flash

This neuron fires on HTML markup tokens—especially Bootstrap’s grid structures, i.e. `<div>` tags with class names like “container,” “row,” and “col--.”

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

_period

-0.07

-Semitic

-0.06

!=

-0.06

:";
↵

-0.06

Pun

-0.06

번

-0.06

.databinding

-0.06

);
↵

-0.06

 telescope

-0.06

 discriminatory

-0.06

POSITIVE LOGITS

 nobody

0.07

 sorts

0.06

alte

0.06

FUL

0.06

ellan

0.06

Ley

0.06

 있어서

0.06

 explosive

0.06

lates

0.06

λό

0.06

Activations Density 0.009%

col

This neuron fires on HTML markup tokens—especially Bootstrap’s grid structures, i.e. `<div>` tags with class names like “container,” “row,” and “col-*-*.”

No Comments

No Known Activations

This neuron fires on HTML markup tokens—especially Bootstrap’s grid structures, i.e. `<div>` tags with class names like “container,” “row,” and “col--.”