INDEX

Explanations

phrases indicating locations

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

describing physical location

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 argint

-0.46

barkeit

-0.40

 Singh

-0.40

 đích

-0.38

Allgemeine

-0.37

PropertyValue

-0.37

̚

-0.36

INSON

-0.36

 some

-0.35

 Party

-0.35

POSITIVE LOGITS

Located

1.55

 Located

1.41

Situated

1.16

 Situated

1.05

 located

0.93

located

0.92

位于

0.90

位於

0.86

 ubicada

0.85

 ubicado

0.82

Activations Density 0.004%

phrases indicating locations

describing physical location

No Comments

No Known Activations