INDEX

Explanations

Sci

np_max-act · gemini-2.0-flash

This neuron detects academic citation strings (e.g. journal abbreviations like “Proc. Natl. Acad. Sci. USA”) 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 bonding

-0.07

_HALF

-0.07

 serotonin

-0.06

 hieronta

-0.06

 Muss

-0.06

(curr

-0.06

амп

-0.06

 haben

-0.06

 prophets

-0.06

 Centro

-0.06

POSITIVE LOGITS

 packet

0.07

you

0.07

?a

0.07

’я

0.06

.Alignment

0.06

Ac

0.06

 alleges

0.06

ैसल

0.06

ș

0.06

 Instructions

0.06

Activations Density 0.002%

Sci

This neuron detects academic citation strings (e.g. journal abbreviations like “Proc. Natl. Acad. Sci. USA”) in the text.

No Comments

No Known Activations