INDEX

Explanations

references to moral and ethical issues surrounding wartime atrocities and animal testing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 connexes

-0.63

abestanden

-0.59

atég

-0.56

 חיצוניים

-0.55

mancy

-0.54

ContextHolder

-0.54

 odkazy

-0.53

 المعرف

-0.52

 propOrder

-0.52

stateProvider

-0.51

POSITIVE LOGITS

 torture

0.87

 cruelty

0.84

 abuse

0.82

 atrocities

0.81

 cruel

0.81

 Torture

0.81

虐

0.79

 tortured

0.79

 tort

0.78

 tortura

0.78

Activations Density 0.379%

references to moral and ethical issues surrounding wartime atrocities and animal testing

No Comments

No Known Activations