INDEX

Explanations

proper nouns, particularly names and geographic locations

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

 ligiloj

-0.98

 Normdatei

-0.90

IUrlHelper

-0.88

aarrggbb

-0.88

 jsPsych

-0.79

 ModelExpression

-0.74

 ***!

-0.73

fromnode

-0.73

 defaultstate

-0.73

 Biôgrafia

-0.72

POSITIVE LOGITS

 instead

0.41

 Spez

0.40

 Stunde

0.38

 least

0.38

 though

0.38

 hamil

0.36

巴巴

0.36

 also

0.36

vesi

0.36

XDECREF

0.35

Activations Density 0.010%

proper nouns, particularly names and geographic locations

No Comments

No Known Activations