INDEX

Explanations

code-related syntax or structured elements within a programming context

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

public methods and functions

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

 protoimpl

-0.75

 gynhyrchwyd

-0.71

➯

-0.70

 verſch

-0.69

Хьажоргаш

-0.67

 queſta

-0.66

Aholisi

-0.66

 geweſen

-0.66

 vooz

-0.65

 zwiſchen

-0.64

POSITIVE LOGITS

GOTREF

0.38

0.35

SuspendLayout

0.32

 Tiago

0.31

 Karl

0.31

ミア

0.30

дні

0.29

public

0.28

US

0.28

Activations Density 0.003%

code-related syntax or structured elements within a programming context

public methods and functions

No Comments

No Known Activations