INDEX

Explanations

This neuron has two modes. It looks for either events in a sexual encounter, but also for certain words in the scientific domain, namely 'sponges', 'silicate', and related things

oai_token-act-pair · gemini-2.0-flash

sponges and related terms

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Sp

-2.11

sp

-2.02

SP

-1.98

Sp

-1.92

SP

-1.77

sp

-1.69

SPR

-1.66

Spo

-1.65

SPA

-1.63

 Spar

-1.62

POSITIVE LOGITS

GraphicsUnit

0.65

Varint

0.62

+#+#

0.61

ClientRect

0.52

IIT

0.50

 OnDestroy

0.49

 Dario

0.49

twimg

0.49

neuf

0.48

IIT

0.48

Activations Density 0.767%

This neuron has two modes. It looks for either events in a sexual encounter, but also for certain words in the scientific domain, namely 'sponges', 'silicate', and related things

sponges and related terms

No Comments

No Known Activations