INDEX

Explanations

terms related to coordinates and spatial relationships

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

 AGENCY

-0.37

 agency

-0.36

Distribuzione

-0.36

 constaté

-0.35

 okuyayım

-0.34

Agency

-0.34

tagHelperRunner

-0.32

 Agency

-0.31

 sounded

-0.30

 edilir

-0.30

POSITIVE LOGITS

 coordinates

0.84

att

0.82

att

0.82

 Coordinates

0.77

Att

0.77

coordinates

0.75

brick

0.75

Coordinate

0.73

 coordinate

0.73

Activations Density 0.999%

terms related to coordinates and spatial relationships

No Comments

No Known Activations