INDEX

Explanations

references to viewpoints or positions in discussions

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

taking a stance

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

CPR

-0.59

autorest

-0.58

 ویکی‌پدی

-0.58

 Infórmanos

-0.56

CDR

-0.55

xFFFFFF

-0.55

']],

-0.51

 otomatig

-0.51

Telefax

-0.51

yyb

-0.50

POSITIVE LOGITS

 stance

1.88

 stances

1.41

stance

1.20

 postura

1.02

 Haltung

0.97

 posture

0.88

stances

0.69

 Stellung

0.68

 Posture

0.67

 actitud

0.64

Activations Density 0.004%

references to viewpoints or positions in discussions

taking a stance

No Comments

No Known Activations

references to viewpoints or positions in discussions

taking a stance

No Comments

No Known Activations