INDEX

Explanations

references to villains and criminal organizations in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 Baillargeon

-0.63

 دیکھیے

-0.59

IsContent

-0.57

ForKey

-0.50

 bint

-0.48

AndEndTag

-0.48

 acar

-0.48

άνει

-0.47

లాలు

-0.47

OON

-0.47

POSITIVE LOGITS

#+#

0.73

 evil

0.72

 OMITBAD

0.61

 terrorist

0.59

IConfiguration

0.58

 SqlCommand

0.58

 indicted

0.58

ⓧ

0.57

reportWebVitals

0.56

Evil

0.56

Activations Density 0.599%

references to villains and criminal organizations in a narrative context

No Comments

No Known Activations