INDEX

Explanations

references to countries involved in geopolitical conflicts, particularly focusing on Iran, North Korea, and Israel

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

__.__

-0.61

newArrayList

-0.59

 الحره

-0.59

pæ

-0.58

евич

-0.54

 pemer

-0.53

RefNanny

-0.52

elsch

-0.50

SCOP

-0.50

llan

-0.49

POSITIVE LOGITS

 Taliban

0.85

 ComVisible

0.79

 Afghan

0.74

WireFormatLite

0.73

 afghan

0.69

setVerticalGroup

0.68

 Afghanistan

0.68

 iran

0.67

MockBean

0.67

 Iran

0.67

Activations Density 0.297%

references to countries involved in geopolitical conflicts, particularly focusing on Iran, North Korea, and Israel

No Comments

No Known Activations