INDEX

Explanations

phrases indicating the location of objects or places

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

"located"

np_acts-logits-general · gemini-2.0-flash

where things are located

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

 wobec

-0.59

genodigd

-0.54

ſelves

-0.53

BASELINE

-0.51

 Einstellung

-0.50

einander

-0.50

 Verſ

-0.49

 myſelf

-0.49

 geſ

-0.49

 despedida

-0.48

POSITIVE LOGITS

 located

1.70

 Located

1.49

 situated

1.48

Located

1.44

located

1.31

 Situated

1.28

Situated

1.21

 situés

1.20

 situé

1.20

 situada

1.16

Activations Density 0.386%

phrases indicating the location of objects or places

"located"

where things are located

No Comments

No Known Activations