INDEX

Explanations

verbs associated with actions and activities

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

verbs followed by prepositions/objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DebuggerNonUser

-0.39

 rest

-0.39

 idén

-0.37

rest

-0.37

 forstå

-0.36

 așa

-0.36

 Rest

-0.36

 plais

-0.36

ArgumentParser

-0.35

__":

-0.35

POSITIVE LOGITS

 الاطلاع

0.64

 tartalomajánló

0.54

səhifə

0.53

HomeAsUpEnabled

0.52

majánló

0.52

addPreferredGap

0.51

$_(

0.50

 myſelf

0.50

erintah

0.50

 cherchés

0.50

Activations Density 0.103%

verbs associated with actions and activities

verbs followed by prepositions/objects

No Comments

No Known Activations