INDEX

Explanations

terms related to terrorism and significant historical attacks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

arim

-0.53

KommentareTeilen

-0.52

ⓧ

-0.51

 bậc

-0.51

seurs

-0.51

geslacht

-0.50

MENAFN

-0.49

onHide

-0.48

ाज

-0.47

ામ

-0.46

POSITIVE LOGITS

 terrorism

0.68

 terrorist

0.65

 незавершена

0.64

 ComVisible

0.61

terror

0.59

 bombings

0.58

Terror

0.57

 terror

0.56

InvalidProtocol

0.56

 >=",

0.56

Activations Density 0.158%

terms related to terrorism and significant historical attacks

No Comments

No Known Activations

terms related to terrorism and significant historical attacks

No Comments

No Known Activations