INDEX

Explanations

references to secretive and dangerous organizations or individuals, particularly those related to crime or assassination

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

Adjectives describing properties or qualities

np_acts-logits-general · gemini-2.0-flash

descriptive qualities or states

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

 <>",

-0.54

lccn

-0.46

InjectAttribute

-0.44

 GetEnumerator

-0.44

 concluding

-0.42

SequentialGroup

-0.42

anterie

-0.41

 ffilm

-0.41

Kesimpulan

-0.40

henswürdigkeiten

-0.39

POSITIVE LOGITS

AutoScaleMode

0.44

Literally

0.40

AddTagHelper

0.40

 ohnehin

0.39

 contextLoads

0.38

Basically

0.37

 getTotal

0.37

 extrema

0.37

 whereof

0.36

totally

0.36

Activations Density 0.497%

references to secretive and dangerous organizations or individuals, particularly those related to crime or assassination

Adjectives describing properties or qualities

descriptive qualities or states

No Comments

No Known Activations