INDEX

Explanations

references to red flags or warning signs

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

red lights and flags

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 AssemblyCompany

-0.39

维护

-0.34

 تضيفلها

-0.33

 chocolat

-0.33

 włas

-0.32

 disambiguazione

-0.32

 geografia

-0.32

ipedia

-0.31

 somos

-0.31

maintain

-0.30

POSITIVE LOGITS



0.64

posedge

0.60

ValueStyle

0.57

invokeLater

0.56

cedure

0.53

 alert

0.51

Sentinel

0.51

red

0.50

 risk

0.49

 alerts

0.48

Activations Density 0.021%

references to red flags or warning signs

red lights and flags

No Comments

No Known Activations