INDEX

Explanations

key nouns and phrases related to books, narratives, experiences, and societal constructs

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

types of events or entities

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

 antaranya

-0.42

 oídos

-0.42

uș

-0.42

 nasled

-0.41

 aktivitas

-0.41

 invitados

-0.41

 atividade

-0.39

 Relaciones

-0.39

 siguientes

-0.38

 změ

-0.38

POSITIVE LOGITS

 worthy

0.54

 worth

0.51

providedIn

0.49

bkz

0.47

 StatelessWidget

0.47

ReusableCell

0.47

stoy

0.46

 very

0.46

 Rumuni

0.45

就是一个

0.45

Activations Density 0.038%

key nouns and phrases related to books, narratives, experiences, and societal constructs

types of events or entities

No Comments

No Known Activations