INDEX

Explanations

numerical values, particularly those related to configurations or settings

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

digits following format specifiers

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

JspWriter

-0.59

elemField

-0.54

 ""],

-0.51

 AspNetCore

-0.51

InputBorder

-0.51

 roleId

-0.50

AndroidJUnit

-0.49

 oprot

-0.49

prepareStatement

-0.49

HandlerContext

-0.48

POSITIVE LOGITS

Referințe

0.44

 menulis

0.43

 Pilihan

0.39

Conservation

0.39

 nakalista

0.39

脚注

0.38

cry

0.38

 Maß

0.38

 yazı

0.37

divia

0.37

Activations Density 0.002%

numerical values, particularly those related to configurations or settings

digits following format specifiers

No Comments

No Known Activations