INDEX

Explanations

references to variables or symbols related to mathematical or data-oriented contexts

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

Strings of two or three letters ending in "z"

np_acts-logits-general · gemini-2.0-flash

az, iz, ez, oz endings

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 bowiem

-0.48

 gorą

-0.47

 chaude

-0.46

 Cuer

-0.45

 wewnętr

-0.44

 těch

-0.43

 obowią

-0.43

 Seil

-0.42

 národ

-0.42

<h1>

-0.42

POSITIVE LOGITS

az

0.95

Taz

0.93

iz

0.90

ez

0.90

zz

0.90

0.88

oz

0.88

Kiz

0.87

Taz

0.87

azy

0.86

Activations Density 0.958%

references to variables or symbols related to mathematical or data-oriented contexts

Strings of two or three letters ending in "z"

az, iz, ez, oz endings

No Comments

No Known Activations