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-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

我が家の

-0.55

ConstraintMaker

-0.51

avax

-0.50

 alcuna

-0.49

laught

-0.49

 GenerationType

-0.48

 trasparente

-0.47

freies

-0.47

 Ignite

-0.47

ワイイ

-0.46

POSITIVE LOGITS

People

1.57

people

1.55

 people

1.52

 People

1.50

 PEOPLE

1.46

PEOPLE

1.41

ppl

1.02

 peop

0.96

 peoples

0.90

 Menschen

0.90

Activations Density 0.099%

references to people

No Comments

No Known Activations