INDEX

Explanations

mentions of individual people and their contributions in various contexts

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

people's names

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Rüyada

-0.40

 incidents

-0.38

więcej

-0.37

 hutan

-0.37

ClInclude

-0.35

WriteTagHelper

-0.35

tanleria

-0.35

 lichaam

-0.34

ernacle

-0.34

Distribuzione

-0.34

POSITIVE LOGITS

setVerticalGroup

0.63

 '\\;'

0.56

 reviewer

0.48

 staffers

0.48

 editors

0.48

 staff

0.48

GTCX

0.46

 OMITBAD

0.45

 editor

0.45

 appreci

0.45

Activations Density 0.024%

mentions of individual people and their contributions in various contexts

people's names

No Comments

No Known Activations