INDEX

Explanations

statements about existence, evidence, and claims related to opinions or facts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.27.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

Bul

-0.48

neman

-0.47

 ferr

-0.47

Bas

-0.46

HideFlags

-0.45

Hochspringen

-0.45

bul

-0.44

Bul

-0.43

 Metropol

-0.43

bas

-0.43

POSITIVE LOGITS

ScopeManager

0.50

 Bewußt

0.48

 koszulka

0.47

 bluzka

0.46

 Grüsse

0.44

 święta

0.41

 ragioni

0.41

 trekken

0.41

 gesteld

0.41

 Flucht

0.40

Activations Density 0.262%

statements about existence, evidence, and claims related to opinions or facts

No Comments

No Known Activations