INDEX

Explanations

references to the color brown, particularly in relation to objects and descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Fontaine

-0.91

Tuk

-0.84

Personensuche

-0.82

YPTO

-0.81

 ostavi

-0.80

 Champlain

-0.79

 Biôgrafia

-0.79

 Minaj

-0.79

 Alfaro

-0.78

*/;

-0.78

POSITIVE LOGITS

 Brown

1.11

Brown

1.08

BROWN

1.02

 BROWN

0.95

 browns

0.93

 brown

0.92

 Browns

0.90

brown

0.83

Jones

0.79

tan

0.79

Activations Density 0.176%

references to the color brown, particularly in relation to objects and descriptions

No Comments

No Known Activations