INDEX

Explanations

references to conflicts involving accusations or investigations related to political figures

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 aceptas

-0.56

出版年

-0.55

 respecte

-0.52

érance

-0.51

uarts

-0.50

rzej

-0.50

 tauto

-0.50

deau

-0.50

épreuve

-0.48

 souver

-0.48

POSITIVE LOGITS

 plot

1.37

 plotting

1.36

 conspiracy

1.28

 plots

1.27

plot

1.24

 Plot

1.23

 conspired

1.21

 conspir

1.17

 Plots

1.16

 plotted

1.16

Activations Density 0.573%

references to conflicts involving accusations or investigations related to political figures

No Comments

No Known Activations