INDEX

Explanations

specific brand names and technical terminology related to products and solutions

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

 ویکی‌پدیا

-0.56

tonode

-0.47

 useAppContext

-0.42

الدراسه

-0.41

Lähteet

-0.40

曖昧さ回避

-0.40

Materiál

-0.40

 Inscrivez

-0.40

 sanitarios

-0.39

ніципалі

-0.39

POSITIVE LOGITS

addGap

0.49



0.44

SharedDtor

0.41

reactstrap

0.41

 समीक्षक

0.40

delwed

0.39

évaluateur

0.38

ılı

0.37

 wisely

0.37

懋

0.37

Activations Density 0.027%

specific brand names and technical terminology related to products and solutions

No Comments

No Known Activations