INDEX

Explanations

references to violent actions and military conflicts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ArgsConstructor

-0.64

ologues

-0.55

 FontWeight

-0.54

יצוני

-0.54

Fprintf

-0.53

strerror

-0.52

 kasarigan

-0.50

 alapján

-0.47

 ää

-0.47

 tác

-0.47

POSITIVE LOGITS

 unsuspecting

1.41

 innocent

1.36

 defen

1.30

innocent

1.12

 inocente

1.07

 helpless

1.04

 innoc

0.99

 unarmed

0.98

 targets

0.97

 hapless

0.93

Activations Density 0.502%

references to violent actions and military conflicts

No Comments

No Known Activations