INDEX

Explanations

specific phrases or constructs indicating a moving or evolving process

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

multilingual words and phrases

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

rungsseite

-0.60

casted

-0.59

 CanadaChoose

-0.47

 attire

-0.47

 respective

-0.46

 disambiguazione

-0.45

 Zudem

-0.44

 Zeitraum

-0.44

 exclu

-0.44

 somit

-0.43

POSITIVE LOGITS

 виправивши

0.56

berdayakan

0.54

 ")");

0.54

 picioare

0.49

 חיצוניים

0.49

というわけで

0.43

 commonest

0.43

 HasFactory

0.43

tinyos

0.42

 allerlei

0.41

Activations Density 0.613%

specific phrases or constructs indicating a moving or evolving process

multilingual words and phrases

No Comments

No Known Activations