INDEX

Explanations

references to figures and citations within a document

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

references to figures and tables

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

 חיצוניים

-0.60

 autorytatywna

-0.58

出版年

-0.55

 rhestr

-0.50

 queſta

-0.48

Spoljašnje

-0.47

Попис

-0.47

 بيها

-0.46

 שוליים

-0.46

EDEFAULT

-0.45

POSITIVE LOGITS

 nakalista

0.59

AccessorTable

0.44

Selanjutnya

0.39

новниш

0.38

 Seitz

0.37

endcsname

0.36

ajuku

0.36

uhkan

0.36

!("{}",

0.36

PrivateRoute

0.35

Activations Density 0.402%

references to figures and citations within a document

references to figures and tables

No Comments

No Known Activations