INDEX

Explanations

comments and annotations in code

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

code comments and keywords

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

Negative Logits

 AssemblyVersion

-0.42

fromnode

-0.42

vetti

-0.41

esistenza

-0.40

 personality

-0.39

Quien

-0.38

.*")]

-0.37

 Krause

-0.36

 nacido

-0.36

 versed

-0.36

POSITIVE LOGITS

 esternos

0.59

 ligiloj

0.57

 lizenzfreies

0.56

 queſta

0.55

 ſind

0.55

Становништво

0.53

Tembelea

0.53

ésultats

0.52

 kasarigan

0.51

 anſ

0.50

Activations Density 0.002%

comments and annotations in code

code comments and keywords

No Known Activations