INDEX

Explanations

names of individuals and their titles or roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_7/width_16k/average_l0_92

Prompts (Dashboard)

24,576 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

ftagPool

-0.42

 EconPapers

-0.36

 vícti

-0.36

vní

-0.36

Gambas

-0.35

 مغ

-0.35

Gilla

-0.35

хьтан

-0.34

서는

-0.34

 intStringLen

-0.31

POSITIVE LOGITS

ValueStyle

0.65

يميديا

0.58

JpaRepository

0.54

 tartalomajánló

0.53

 Administrativna

0.53

IsMutable

0.52

FunctionFlags

0.50

">—

0.49

MessageTagHelper

0.49

titleMargin

0.49

Activations Density 0.887%

names of individuals and their titles or roles

No Comments

No Known Activations