INDEX

Explanations

equality

np_max-act · gemini-2.0-flash

The neuron fires on equality/comparison operator tokens (e.g. “=”, “==”) and related words like “equal” or “not equal.”

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

comment

-0.06

 Brew

-0.06

(pre

-0.06

_DATE

-0.06

(Tag

-0.06

 troubled

-0.06

 nhưng

-0.06

 -------------------------------------------------------------------------↵

-0.06

_recommend

-0.06

(($

-0.06

POSITIVE LOGITS

kových

0.07

ifferent

0.07

 lãi

0.07

ismatic

0.07

dbo

0.06

_ud

0.06

 прот

0.06

разд

0.06

Equal

0.06

 jiného

0.06

Activations Density 0.012%

equality

The neuron fires on equality/comparison operator tokens (e.g. “=”, “==”) and related words like “equal” or “not equal.”

No Comments

No Known Activations