INDEX

Explanations

references to licensing and copyright terms in the text

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

agreed to

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

-0.44

 added

-0.40

 apsau

-0.40

 actionPerformed

-0.39

 Barbier

-0.38

 practised

-0.38

 uovo

-0.37

InputTagHelper

-0.37

 cappello

-0.36

 nabíz

-0.36

POSITIVE LOGITS

 AssemblyCompany

0.77

出版年

0.58

parsedMessage

0.57

 समीक्षाओं

0.52

+#+

0.49

WindowConstants

0.46

 ComVisible

0.45

 estekak

0.45

انگلیسی

0.45

omitempty

0.45

Activations Density 0.001%

references to licensing and copyright terms in the text

agreed to

No Comments

No Known Activations