INDEX

Explanations

references to specific character labels or identifiers in a structured format

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

numbers followed by letters

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

\{\\

-0.47

 guste

-0.42

 juridiques

-0.41

endsection

-0.38

 Leder

-0.37

 zelve

-0.35

 complètes

-0.35

 dichiarato

-0.35

 Freiheit

-0.34

 henvisninger

-0.33

POSITIVE LOGITS

RectangleBorder

0.73

 '\\;'

0.70

NewUrlParser

0.62

Hochspringen

0.61

+#+#

0.58

المشاركات

0.56

 estekak

0.56

tagHelperRunner

0.53

 Walkover

0.53

Билгалдахарш

0.53

Activations Density 0.103%

references to specific character labels or identifiers in a structured format

numbers followed by letters

No Comments

No Known Activations