INDEX

Explanations

terms related to the concept of 'enemy' or adversaries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 Unterhaltung

-0.59

KindOfClass

-0.54

spill

-0.52

MessageOf

-0.52

LookAnd

-0.47

ringe

-0.47

titleMargin

-0.47

="@+

-0.46

 препратки

-0.46

 AssemblyCompany

-0.45

POSITIVE LOGITS

 enemy

0.75

 inimigo

0.69

enemy

0.65

 nemici

0.62

 enemigo

0.62

 Active

0.60

 nemico

0.59

 Enemy

0.59

active

0.58

 ally

0.58

Activations Density 0.202%

terms related to the concept of 'enemy' or adversaries

No Comments

No Known Activations

terms related to the concept of 'enemy' or adversaries

No Comments

No Known Activations