INDEX

Explanations

references to books and publications

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

author, advocate, expert, book

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ContentValues

-0.56

 beginnetje

-0.45

ConverterFactory

-0.45

 hochzeit

-0.45

IntoConstraints

-0.44

IsContent

-0.43

 Photocase

-0.43

 otomatig

-0.43

sillos

-0.43

WebControls

-0.43

POSITIVE LOGITS

новништво

0.42

BeginContext

0.40

тели

0.37

)‏

0.35

vikle

0.35

kiyor

0.35

Parcelize

0.35

lös

0.34

 ilmu

0.34

Activations Density 0.033%

references to books and publications

author, advocate, expert, book

No Comments

No Known Activations