INDEX

Explanations

code-related keywords and syntax structures

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

 Tatsache

-0.45

 przet

-0.43

 feit

-0.42

RTSN

-0.42

 protoimpl

-0.41

 Öffentlichkeit

-0.41

 educativas

-0.41

 Wortes

-0.40

 larmes

-0.40

HtmlAttribute

-0.39

POSITIVE LOGITS

endphp

0.52

 betweenstory

0.51

 Légion

0.47

 Савезне

0.47

 StatelessWidget

0.46

AnchorStyles

0.46

 Aven

0.46

ISupport

0.45

 capitol

0.44

 Routine

0.43

Activations Density 0.040%

code-related keywords and syntax structures

No Comments

No Known Activations