INDEX

Explanations

references to political power struggles and manipulations involving influential figures or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

parsedMessage

-0.60

thique

-0.56

AtIndexPath

-0.54

Hochspringen

-0.54

 Balkans

-0.54

Portail

-0.52

 noDo

-0.52

JvmStatic

-0.51

<bos>

-0.50

Manbalar

-0.49

POSITIVE LOGITS

 foreign

0.90

 British

0.90

 asing

0.88

 outsiders

0.88

 foreigners

0.85

British

0.77

 فريبيس

0.77

foreign

0.76

 foreigner

0.73

 Western

0.72

Activations Density 0.386%

references to political power struggles and manipulations involving influential figures or groups

No Comments

No Known Activations