INDEX

Explanations

questions and conditional statements within code

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

code completion and structure

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

دانشنامهٔ

-0.64

iastical

-0.50

__*/

-0.45

AutoScale

-0.45

 שוליים

-0.43

thums

-0.43

⋙

-0.42

なる

-0.41

 समीक्षाएं

-0.40

 HasFactory

-0.40

POSITIVE LOGITS

bildēt

0.49

middels

0.49

 âmes

0.48

{?

0.45

 tež

0.45

>?

0.44

%?

0.43

()?.

0.43

opardy

0.43

Activations Density 0.003%

questions and conditional statements within code

code completion and structure

No Comments

No Known Activations