INDEX

Explanations

references to people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_14/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

我が家の

-0.53

avax

-0.48

 sinistro

-0.43

 محفوظة

-0.43

glBind

-0.41

 aisladas

-0.40

 coroa

-0.39

 Clovis

-0.38

 indipendente

-0.38

wußt

-0.38

POSITIVE LOGITS

People

1.41

 People

1.38

people

1.34

PEOPLE

1.32

 PEOPLE

1.31

 people

1.23

 Peoples

0.93

 peop

0.92

ppl

0.91

 peoples

0.89

Activations Density 0.065%

references to people

No Comments

No Known Activations