INDEX

Explanations

references to authors and their literary works

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

verbs followed by statements or actions

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

GEBURTSDATUM

-0.60

 ProtoMessage

-0.56



-0.54

KommentareTeilen

-0.53

 newBuilder

-0.52

StoryboardSegue

-0.51

IContainer

-0.50

 touristique

-0.48

lusconi

-0.48



-0.45

POSITIVE LOGITS

 transformative

0.53

 feminist

0.49

 activism

0.48

 activist

0.48

 impactful

0.47

 queer

0.47

 nuanced

0.47

 collaboratively

0.46

 holistic

0.45

 LGBTQ

0.45

Activations Density 0.024%

references to authors and their literary works

verbs followed by statements or actions

No Comments

No Known Activations