INDEX

Explanations

words and phrases indicating political decisions and actions related to treaties and agreements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

 leveraging

-0.60

脚注の使い方

-0.56

 fermés

-0.56

 déterminé

-0.54

 standalone

-0.54

 africains

-0.54

 similaire

-0.54

 prioritizing

-0.54

 Drives

-0.53

 informée

-0.53

POSITIVE LOGITS

piecze

0.70

 altogether

0.65

basicConfig

0.62

kháu

0.59

utilisons

0.59

 تضيفلها

0.58

SourceChecksum

0.58

 AssemblyCompany

0.57

 protoimpl

0.57

 doubtless

0.55

Activations Density 0.661%

words and phrases indicating political decisions and actions related to treaties and agreements

No Comments

No Known Activations