INDEX

Explanations

architectural features and design elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

évaluateur

-0.63

ContentAlignment

-0.62

 tartalomajánló

-0.61

 Agra

-0.61

arakhand

-0.61

cticide

-0.61

 Shaksp

-0.59

'{@

-0.59

 Confucian

-0.59

$_['

-0.56

POSITIVE LOGITS

 design

1.16

 modernist

1.07

 designs

1.05

 Modern

1.03

 modern

1.03

 style

0.97

design

0.97

modern

0.96

 Design

0.96

Modern

0.95

Activations Density 0.185%

architectural features and design elements

No Comments

No Known Activations