INDEX

Explanations

sentences that contain statements or assertions

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

these results indicate

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

DockStyle

-0.62

ValueStyle

-0.61

RenderAtEndOf

-0.60

 unknownFields

-0.57

 мәкал

-0.56

 geſch

-0.55

 GoogleFonts

-0.55

\{\\

-0.54

 étoit

-0.52

㤵

-0.52

POSITIVE LOGITS

ponses

0.32

 Walkover

0.32

AnchorStyles

0.32

urtstag

0.30

олові

0.29

Viitteet

0.29

 pieces

0.29

ępo

0.28

 отвечает

0.28

SpringRunner

0.28

Activations Density 0.210%

sentences that contain statements or assertions

these results indicate

No Comments

No Known Activations