INDEX

Explanations

statements or declarations made by individuals, particularly those in authority or leadership positions

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

announcements by officials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

-0.48

jspb

-0.41

 disambiguazione

-0.40

 Laboratory

-0.39

SharedCtor

-0.38

 Elisha

-0.36

Personendaten

-0.36

ungkin

-0.35

 laboratories

-0.35

 мәкал

-0.35

POSITIVE LOGITS

TypedDataSet

0.59

 announced

0.50

帖最后由

0.47

announced

0.46

 HttpNotFound

0.45

parsedMessage

0.42

 AssemblyVersion

0.41

setVerticalGroup

0.41

 announcement

0.40

 lópez

0.40

Activations Density 0.008%

statements or declarations made by individuals, particularly those in authority or leadership positions

announcements by officials

No Comments

No Known Activations