INDEX

Explanations

phrases indicating high levels or heightened risks

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

Relatively large quantities or degrees

np_acts-logits-general · gemini-2.0-flash

high levels, rates, or amounts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

yelidikan

-0.53

 saisi

-0.50

 appContext

-0.49

ión

-0.48

Props

-0.48

 stronę

-0.47

 encantado

-0.46

 prostředí

-0.46

 Espace

-0.46

 Atlántico

-0.45

POSITIVE LOGITS

 levels

0.95

WriteAttribute

0.85

 HIGH

0.83

 rates

0.77

 level

0.76

 High

0.75

 high

0.75

levels

0.75

High

0.74

 LEVELS

0.73

Activations Density 0.192%

phrases indicating high levels or heightened risks

Relatively large quantities or degrees

high levels, rates, or amounts

No Comments

No Known Activations