INDEX

Explanations

instances of exposure to potentially harmful substances or conditions

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

Exposure or contact with something harmful

np_acts-logits-general · gemini-2.0-flash

exposure

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



-0.73

MemoryWarning

-0.67

 JpaRepository

-0.64

DllImport

-0.64

 Masson

-0.62

stateProvider

-0.60

zarchiwizowane

-0.59

HomeAsUpEnabled

-0.56

gonic

-0.56

 CoreData

-0.56

POSITIVE LOGITS

 exposure

2.45

 Exposure

2.16

exposure

2.09

Exposure

2.01

 exposed

1.93

exposed

1.82

 EXPOSURE

1.80

 exposures

1.77

 Exposed

1.67

 expose

1.63

Activations Density 0.430%

instances of exposure to potentially harmful substances or conditions

Exposure or contact with something harmful

exposure

No Comments

No Known Activations