INDEX

Explanations

phrases that express causation or reasons

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

Immediately precedes "the" and follows "to" or "of"

np_acts-logits-general · gemini-2.0-flash

reasons for consequences

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uxxxx

-0.57

 Athenians

-0.57

 Allegretto

-0.57

 Agamemnon

-0.53

 itſelf

-0.51

 idéia

-0.51

 persino

-0.50

 sendiri

-0.50

 myſelf

-0.49

 ་་

-0.49

POSITIVE LOGITS

 lack

0.90

 reasons

0.85

its

0.81

 their

0.75

reasons

0.73

 limitations

0.72

 adanya

0.71

 being

0.71

 necessity

0.70

 differences

0.69

Activations Density 0.206%

phrases that express causation or reasons

Immediately precedes "the" and follows "to" or "of"

reasons for consequences

No Comments

No Known Activations