INDEX

Explanations

instances of the word "complicated" and related terms emphasizing complexity

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

explaining complexity

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

 său

-0.61

ždy

-0.49

 jäl

-0.49

 orejas

-0.48

 pères

-0.47

 beliau

-0.47

 dumne

-0.47

 Autorin

-0.47

 żyw

-0.47

NOPQRST

-0.45

POSITIVE LOGITS

complicated

1.08

 complicated

1.00

 complica

0.82

 complications

0.79

 complicate

0.78

trouble

0.77

 complication

0.75

licated

0.75

 complic

0.73

XmlType

0.71

Activations Density 0.005%

instances of the word "complicated" and related terms emphasizing complexity

explaining complexity

No Comments

No Known Activations