INDEX

Explanations

references to programming constructs and code elements

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

IP protocol codes

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

 reina

-0.37

 tölt

-0.35

Во

-0.35

Transkript

-0.34

Pump

-0.33

Stand

-0.33

 snippetHide

-0.33

warnai

-0.33

badan

-0.33

 frame

-0.33

POSITIVE LOGITS

 ویکی‌آمباردا

0.66

ThroughAttribute

0.57

 propOrder

0.56

 JpaRepository

0.55

 Audiodateien

0.54

الدراسه

0.54



0.54

RenderAtEndOf

0.54

Obrázky

0.52

CppMethod

0.52

Activations Density 0.069%

references to programming constructs and code elements

IP protocol codes

No Comments

No Known Activations

references to programming constructs and code elements

IP protocol codes

No Comments

No Known Activations