INDEX

Explanations

numerical identifiers or labels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 noel

-0.66

 fers

-0.64

Saman

-0.62

teng

-0.62

 exper

-0.60

 abel

-0.59

nowrap

-0.58

ferd

-0.58

 pascal

-0.58

 Balth

-0.57

POSITIVE LOGITS

ii

2.20

iii

1.87

ii

1.81

iii

1.52

iiiii

1.41

iiii

1.37

vii

1.33

sii

1.31

viii

1.25

Vii

1.17

Activations Density 0.188%

numerical identifiers or labels

No Comments

No Known Activations