INDEX

Explanations

mentions of various professional roles and groups of people

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

groups of people

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

StoryboardSegue

-0.39

Filmographie

-0.34

 chimiques

-0.32

 nakalista

-0.32

richTextPanel

-0.31

 surla

-0.31

 cartera

-0.30

tit

-0.30

 labour

-0.29

?}",

-0.29

POSITIVE LOGITS

 alike

0.63

stdc

0.60

protoimpl

0.59

 EconPapers

0.58

ResponseWriter

0.57

AntiForgeryToken

0.55

ſelf

0.51

ocrats

0.51

Everyone

0.50

 setempat

0.50

Activations Density 0.031%

mentions of various professional roles and groups of people

groups of people

No Comments

No Known Activations