INDEX

Explanations

names and entities relevant to legal cases and medical conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 InputDecoration

-0.90

 Leinwand

-0.86

 Teig

-0.83

 Zwiebel

-0.82

 Baus

-0.81



-0.80

 Kuchen

-0.78

 Ström

-0.77

 Futter

-0.75

 Sucher

-0.75

POSITIVE LOGITS

 Vettel

0.41

 Jock

0.41

 Tuchel

0.37

sApp

0.35

jenige

0.33

 Muck

0.33

Schmidt

0.33

Horn

0.32

 választ

0.32

 Otto

0.31

Activations Density 1.007%

names and entities relevant to legal cases and medical conditions

No Comments

No Known Activations