INDEX

Explanations

words that describe other people, especially coworkers and teammates

oai_token-act-pair · gemini-2.0-flash

coworkers and neighbors

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_8/width_16k/average_l0_52

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.8.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 teammate

-0.75

 للمعارف

-0.75

 colleague

-0.74

 Neighbor

-0.69

neighbor

-0.68

 comrade

-0.68

neighbour

-0.68

 neighbor

-0.67

 mates

-0.66

devamını

-0.66

POSITIVE LOGITS

enschappelijke

0.48

 '\\;'

0.47

IContainer

0.45

riedenheit

0.43

nationality

0.42

imanapun

0.41

 quanto

0.41

 članak

0.40

NamedQuery

0.39

 Reverso

0.38

Activations Density 4.340%

words that describe other people, especially coworkers and teammates

coworkers and neighbors

No Comments

No Known Activations