INDEX

Explanations

mentions of "thing" or "things."

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

the word thing

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

 dianteira

-0.55

 térmico

-0.54

 termica

-0.48

 completas

-0.47

 Gestão

-0.46

 partenza

-0.46

 enfans

-0.45

 doméstica

-0.45

 financeiro

-0.45

 créditos

-0.44

POSITIVE LOGITS

 what

0.63

 things

0.60

ThroughAttribute

0.59

Things

0.58

 Things

0.57

 THINGS

0.57

LabelTagHelper

0.56

ölkerung

0.55

things

0.54

 whats

0.54

Activations Density 0.019%

mentions of "thing" or "things."

the word thing

No Comments

No Known Activations