INDEX

Explanations

terms related to permissions and settings management

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

Returning to a prior state

np_acts-logits-general · gemini-2.0-flash

returning to baseline

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 Wikidata

-0.64

AndEndTag

-0.63

 Signalez

-0.59

ittäin

-0.59

PerformLayout

-0.59

 EClass

-0.58

__(/*!

-0.57

 LayoutInflater

-0.56

 ExecuteAsync

-0.56

pushFollow

-0.55

POSITIVE LOGITS

 restored

1.63

 restore

1.52

 restoring

1.52

 restores

1.46

 reverted

1.40

 reinstated

1.32

restore

1.30

 restoration

1.30

 Restore

1.29

 reverting

1.28

Activations Density 0.418%

terms related to permissions and settings management

Returning to a prior state

returning to baseline

No Comments

No Known Activations