INDEX

Explanations

references to software usage, updates, or version changes

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

Preceding a colon or bracket

np_acts-logits-general · gemini-2.0-flash

provides, cancels, becomes, adds, to

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

 also

-0.47

and

-0.47

 however

-0.43

(«

-0.40

 Réponses

-0.40

also

-0.39

 heck

-0.39

(£

-0.39

給我

-0.38

הת

-0.38

POSITIVE LOGITS

rungsseite

0.90

GEBURTSDATUM

0.84

 AssemblyTitle

0.75

addCriterion

0.71

 وتسجيلات

0.69

 있어

0.68

kháu

0.68

 متعلقه

0.65

;';

0.63

弭

0.63

Activations Density 0.032%

references to software usage, updates, or version changes

Preceding a colon or bracket

provides, cancels, becomes, adds, to

No Comments

No Known Activations