INDEX

Explanations

references to division or categorization in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

ופת

-0.62

<em>

-0.59

Duc

-0.59

 Piston

-0.59

évaluateur

-0.58

 pistons

-0.58

 Regency

-0.57

 bson

-0.57

ImageButton

-0.57

szön

-0.57

POSITIVE LOGITS

 divide

1.33

 divided

1.33

 divides

1.29

 Divided

1.27

 dividing

1.25

divided

1.25

 Divide

1.14

divide

1.13

 Separation

1.04

 divid

1.04

Activations Density 0.148%

references to division or categorization in various contexts

No Comments

No Known Activations