INDEX

Explanations

numerical values and references to legal or regulatory contexts

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

 CanadaChoose

-0.53

ocumented

-0.51

 procedura

-0.49

 otim

-0.48

vedades

-0.48

modity

-0.47

ppuden

-0.47

featureID

-0.47

 Involved

-0.46

醐

-0.45

POSITIVE LOGITS

 transformative

0.40

“

0.36

GND

0.36

 passim

0.36

 brief

0.35

 argues

0.35

ParseTree

0.35

 debate

0.34

djangoproject

0.33

 otomatig

0.33

Activations Density 0.001%

numerical values and references to legal or regulatory contexts

No Comments

No Known Activations