INDEX

Explanations

programming-related concepts and terms

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

code constructs

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

下载附件

-0.47

awtextra

-0.46

CppMethod

-0.42

NUMX

-0.42

 JpaRepository

-0.40

𓃵

-0.40

+#+

-0.39

principalColumn

-0.39

 typelib

-0.39

ftagPool

-0.38

POSITIVE LOGITS

ArgsConstructor

0.48

الحياه

0.47

 समीक्षक

0.43

instancetype

0.42

 OMITBAD

0.41

anyahu

0.40

raisemb

0.38

 Emeritus

0.38

JMenu

0.37

JMenuItem

0.37

Activations Density 0.006%

programming-related concepts and terms

code constructs

No Comments

No Known Activations