INDEX

Explanations

numerical identifiers or markers of significance

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

git diff @@

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

.*")]

-0.47

TestingModule

-0.47

getSystemService

-0.44

RequestMapping

-0.43

ExtendWith

-0.43

IActionResult

-0.43

SerializeField

-0.41

NameValuePair

-0.40

AnimationsModule

-0.40

newArrayList

-0.40

POSITIVE LOGITS

AddColumn

1.06

 autorytatywna

0.53

DropColumn

0.52

BeginContext

0.51

 незавершена

0.50

 فريبيس

0.49

httphttps

0.45

Handlung

0.45

uxxxx

0.45

 ſind

0.44

Activations Density 0.000%

numerical identifiers or markers of significance

git diff @@

No Comments

No Known Activations