INDEX

Explanations

phrases related to deception and manipulation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

OGND

-0.45

 autorytatywna

-0.35

 sécher

-0.32

 esperanza

-0.30

 Slaughter

-0.28

 symp

-0.28

ttp

-0.28

пры

-0.28

__(/*!

-0.28

espoir

-0.27

POSITIVE LOGITS

 report

0.71

ReusableCell

0.66

report

0.65

 Report

0.63

報告

0.62

 REPORT

0.62

 &___

0.61

汇报

0.60

Report

0.60

 presentation

0.58

Activations Density 0.337%

phrases related to deception and manipulation

No Comments

No Known Activations