INDEX

Explanations

references to original works, including quotations, citations, and authorship

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Emphasis on "original"

np_acts-logits-general · gemini-2.0-flash

emphasis in original

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RenderAtEndOf

-1.02

hidupan

-0.64

complexContent

-0.60

 propOrder

-0.57

setVerticalGroup

-0.56

popd

-0.54

abestanden

-0.54

mampuan

-0.52

bewerken

-0.51



-0.50

POSITIVE LOGITS

 original

3.74

original

3.39

Original

3.14

 Original

3.00

 ORIGINAL

2.88

 originals

2.82

 originales

2.74

ORIGINAL

2.74

 originali

2.65

 originale

2.65

Activations Density 0.615%

references to original works, including quotations, citations, and authorship

Emphasis on "original"

emphasis in original

No Comments

No Known Activations