INDEX

Explanations

usage of HTML or code syntax elements

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

Constructor and Description

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

 HasFactory

-0.54

########.

-0.52

tidaknya

-0.52

RegressionTest

-0.52

 propOrder

-0.52

eira

-0.49

帖最后由

-0.48

rungsseite

-0.47

KURZBESCHREIBUNG

-0.47

 дописавши

-0.47

POSITIVE LOGITS

ftagPool

0.40

 accord

0.38

̍t

0.36

twimg

0.36

col

0.34

gered

0.34

SpringRunner

0.33

LTRB

0.33

tierrez

0.33

AntiForgeryToken

0.33

Activations Density 0.001%

usage of HTML or code syntax elements

Constructor and Description

No Comments

No Known Activations