INDEX

Explanations

punctuations

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

 للمعارف

-0.61

 kasarigan

-0.50

 <<<<<<<<<<<<<<

-0.47

httphttps

-0.46

findpost

-0.45

出版年

-0.43

ResumeLayout

-0.40

 Dicht

-0.40

rzost

-0.39

 disambiguazione

-0.39

POSITIVE LOGITS

 ExecuteAsync

0.46

ChildScrollView

0.45

 three

0.45

 bullion

0.44

0.43

탤

0.43

ласти

0.42

 four

0.42

ManyToOne

0.41

Activations Density 0.100%

punctuations

No Comments

No Known Activations