INDEX

Explanations

symbols and formatting commonly used in academic citations or references

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

doi numbers and identifiers

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

GEBURTSDATUM

-0.47

 Савезне

-0.47

 Administrativna

-0.47

 otomatig

-0.45

LocalDateTime

-0.41

 stří

-0.39

lík

-0.39

bildeten

-0.39

giant

-0.38

Portail

-0.38

POSITIVE LOGITS

thâu

0.54

SequentialGroup

0.48

出版年

0.43

WebControls

0.41

 Normdatei

0.40

+#+#

0.40

esModule

0.40

localctx

0.38

 authorized

0.37

antası

0.37

Activations Density 0.066%

symbols and formatting commonly used in academic citations or references

doi numbers and identifiers

No Comments

No Known Activations