INDEX

Explanations

lines of code or syntax structures in programming contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

end of code block

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.32.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

 برانيه

-0.51

rungsseite

-0.47

 يتيمه

-0.46

camp

-0.43

mical

-0.42

mated

-0.42

cor

-0.42

 Mena

-0.42

pore

-0.41

engers

-0.41

POSITIVE LOGITS

MemoryWarning

0.50

 snippetHide

0.48

 celebrid

0.48

 pungkasnya

0.47

endpush

0.47

pecabe

0.47

 celebridades

0.47

 pungkas

0.46

 senare

0.43

 sostenibilidad

0.43

Activations Density 0.006%

lines of code or syntax structures in programming contexts

end of code block

No Comments

No Known Activations