INDEX

Explanations

names and email addresses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 Chwiliwch

-1.20

帖最后由

-1.16

 kasarigan

-1.14

DockStyle

-1.10

 كومونز

-1.06

 ModelExpression

-1.05

 nakalista

-1.04



-1.02

 surla

-1.01

 propOrder

-1.01

POSITIVE LOGITS

0.34

主

0.30

णी

0.29

0.27

now

0.26

gmail

0.26

 agora

0.25

values

0.25

лён

0.24

Activations Density 0.326%

names and email addresses

No Comments

No Known Activations