INDEX

Explanations

references to varying types of materials and their classifications or identifiers

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

Two-letter sequences starting with "Z"

np_acts-logits-general · gemini-2.0-flash

z-ending words and sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

 прави

-0.40

 something

-0.38

 normally

-0.37

 richtigen

-0.37

 richtige

-0.37

 paik

-0.37

 národ

-0.36

 inneren

-0.36

tranno

-0.35

MarshalTo

-0.35

POSITIVE LOGITS

BZ

0.65

zing

0.63

BZ

0.62

KZ

0.62

zes

0.60

 Ozone

0.59

Kiz

0.58

Dez

0.58

zed

0.57

oz

0.57

Activations Density 0.358%

references to varying types of materials and their classifications or identifiers

Two-letter sequences starting with "Z"

z-ending words and sequences

No Comments

No Known Activations