INDEX

Explanations

references to personal belongings or items owned by the speaker

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

"I" followed by a verb

np_acts-logits-general · gemini-2.0-flash

"I" doing things or having preferences

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

setVerticalGroup

-0.59

 bearer

-0.51

 ofrecerte

-0.49

potentially

-0.47

 demain

-0.47

ši

-0.46

 rivi

-0.46

 foreseen

-0.46

 proposing

-0.45

 restauration

-0.45

POSITIVE LOGITS

 usually

1.14

 often

1.14

 brukar

1.01

 always

0.99

 sometimes

0.99

 prefer

0.97

 routinely

0.96

 typically

0.96

 thường

0.94

usually

0.93

Activations Density 0.350%

references to personal belongings or items owned by the speaker

"I" followed by a verb

"I" doing things or having preferences

No Comments

No Known Activations