INDEX

Explanations

references to emergency responses and health-related incidents

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

Someone notifying or contacting someone else

np_acts-logits-general · gemini-2.0-flash

reporting or telling people

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

 vectorielles

-0.54

∮

-0.45

calientes

-0.45

Geografi

-0.44

 Reſ

-0.43

ClientRect

-0.42

RTRS

-0.42

 Anſ

-0.41

Guido

-0.40

 setw

-0.40

POSITIVE LOGITS

 menghubungi

0.56

 ब्रेकडाउन

0.56

 melap

0.54

 contacting

0.54

 contacted

0.52

 notifying

0.52

 call

0.47

immediately

0.45

 consulted

0.44

 consulte

0.44

Activations Density 0.308%

references to emergency responses and health-related incidents

Someone notifying or contacting someone else

reporting or telling people

No Comments

No Known Activations