INDEX

Explanations

references to political figures and their actions

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

suggesting something is unusual or wrong

np_acts-logits-general · gemini-2.0-flash

suspicious or unusual details

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 صوتيه

-0.52

 ***!

-0.45

UnusedPrivate

-0.43

+#+#

-0.42

jspx

-0.42

ValueStyle

-0.41

 occafion

-0.39

Contained

-0.39

henswürdigkeiten

-0.38

 cherchés

-0.38

POSITIVE LOGITS

 suspiciously

0.60

 suspicious

0.50

wußt

0.47

sterious

0.46

MigrationBuilder

0.44

 étoient

0.44

odd

0.43

 soudain

0.42

 demasi

0.42

 misterioso

0.42

Activations Density 0.646%

references to political figures and their actions

suggesting something is unusual or wrong

suspicious or unusual details

No Comments

No Known Activations