INDEX

Explanations

symbols or special characters within written text

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

WireFormatLite

-0.59

addContainerGap

-0.54

 canst

-0.53

ſelf

-0.53

TabIndex

-0.51

 MainAxisSize

-0.51

SharedCtor

-0.51

 оригіналу

-0.50

 Anſ

-0.50

ParallelGroup

-0.50

POSITIVE LOGITS

*__

0.42

Cyfarwyddwr

0.41

 endemic

0.41

 costumi

0.41

::<

0.39

 Japanese

0.39

 parasitic

0.39

 scientist

0.38

 Indian

0.38

 ***!

0.38

Activations Density 0.061%

symbols or special characters within written text

No Comments

No Known Activations