INDEX

Explanations

locations and geography-related descriptors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 fær

-0.42

 onPostExecute

-0.40

entuh

-0.39

 kasarigan

-0.38

vědět

-0.36

참고

-0.36

 Plast

-0.36

 condamné

-0.36

Flu

-0.36

homo

-0.35

POSITIVE LOGITS

CloseOperation

0.64

 located

0.54

 Located

0.53

 downtown

0.53

located

0.53

 sija

0.51

Located

0.48

ⓧ

0.48

 situés

0.47

ondissement

0.46

Activations Density 0.372%

locations and geography-related descriptors

No Comments

No Known Activations

locations and geography-related descriptors

No Comments

No Known Activations