INDEX

Explanations

terms related to door-related actions and thresholds

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

following "door" or "doors"

np_acts-logits-general · gemini-2.0-flash

opening and closing doors

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

ритори

-0.57

Референце

-0.47

tld

-0.45

 EconPapers

-0.43

///</

-0.43

🏻

-0.42

 referenties

-0.42

ený

-0.41

tinyos

-0.41

currentColor

-0.41

POSITIVE LOGITS

 door

2.93

 doors

2.75

door

2.41

 Door

2.31

Door

2.27

 DOOR

2.26

doors

2.17

 puerta

2.15

 gates

2.14

 Doors

2.11

Activations Density 0.188%

terms related to door-related actions and thresholds

following "door" or "doors"

opening and closing doors

No Comments

No Known Activations