INDEX

Explanations

key terms and phrases that signal important concepts or items in a technical context

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

Personendaten

-0.50

Kontrola

-0.49

enumii

-0.48

makeText

-0.48

yntaxException

-0.42

 pinulongan

-0.42

 betweenstory

-0.42

 باخ

-0.41

adaptiveStyles

-0.37

fromnode

-0.37

POSITIVE LOGITS

Билгалдахарш

0.61

 Goose

0.49

Goose

0.46

 للمعارف

0.44

OGND

0.43

 '\\;'

0.42

Cyfeiriadau

0.41

Manbalar

0.41

 giorgio

0.41

postsleuth

0.40

Activations Density 0.003%

key terms and phrases that signal important concepts or items in a technical context

No Comments

No Known Activations