INDEX

Explanations

actions and their implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

mybatisplus

-0.80

tvguidetime

-0.77

 verſch

-0.75

 beſch

-0.75

 ſeinen

-0.75

iffance

-0.75

iſen

-0.74

tiérrez

-0.74

 gyhoeddwyd

-0.74

Identyfik

-0.72

POSITIVE LOGITS

 Keuangan

0.41

 Lingkungan

0.39

Bronnen

0.33

 Masyarakat

0.32

 Semoga

0.31

venidos

0.31

<em>

0.29

ChildScrollView

0.28

 polaire

0.28

 olympique

0.28

Activations Density 0.727%

actions and their implications

No Comments

No Known Activations