INDEX

Explanations

expressions of frustration or demands for silence

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

Followed by conjunctions, commas or periods

np_acts-logits-general · gemini-2.0-flash

stop or quit

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

openSession

-0.58

 ändå

-0.56

 CascadeType

-0.51

 насеље

-0.48

Tapi

-0.48

butuhkan

-0.47

BagConstraints

-0.47

Erreferentziak

-0.46

appan

-0.46

getMenuInflater

-0.45

POSITIVE LOGITS

 Shut

1.15

 stop

1.13

 shut

1.13

Shut

1.08

 Stop

1.05

 SHUT

1.01

stop

1.01

Stop

1.01

shut

0.99

 quit

0.95

Activations Density 0.190%

expressions of frustration or demands for silence

Followed by conjunctions, commas or periods

stop or quit

No Comments

No Known Activations