INDEX

Explanations

phrases related to legal actions, such as arrests, charges, and court proceedings

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

followed by legal or bureaucratic actions/events

np_acts-logits-general · gemini-2.0-flash

decisions and outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

MessageTagHelper

-0.85

SourceChecksum

-0.51

 дописавши

-0.48

farwyddwr

-0.44

 заве

-0.42

맑

-0.41

 alibi

-0.41

 Pogba

-0.41

thansa

-0.41

 maso

-0.40

POSITIVE LOGITS

 creș

0.44

mbggenerated

0.42

0.41

 kandungan

0.40

GeneratedCode

0.40

 rospy

0.40

isSuccessful

0.39

 affinch

0.38

 defaultstate

0.36

ArgumentParser

0.36

Activations Density 0.753%

phrases related to legal actions, such as arrests, charges, and court proceedings

followed by legal or bureaucratic actions/events

decisions and outcomes

No Comments

No Known Activations