INDEX

Explanations

phrases indicating comparisons or contrasts

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 volantes

-0.45

 entendido

-0.37

 muerta

-0.37

 útiles

-0.37

 volna

-0.35

 turma

-0.35

 kolye

-0.35

 conseguido

-0.35

 aprendido

-0.35

 Grüsse

-0.35

POSITIVE LOGITS

хьтан

0.62

 Савезне

0.60

httphttps

0.59

LookAnd

0.56

balleur

0.54

ValueGenerated

0.54

<?

0.54

rungsseite

0.54

 intStringLen

0.54

<",

0.53

Activations Density 0.073%

phrases indicating comparisons or contrasts

No Comments

No Known Activations