INDEX

Explanations

code structure and syntax within programming or scripting contexts

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

documentation and code 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

 betweenstory

-0.94

 propOrder

-0.88

RenderAtEndOf

-0.81

 autorytatywna

-0.80

transQ

-0.78

Datuak

-0.78

MigrationBuilder

-0.76

ConstraintMaker

-0.72

[@BOS@]

-0.71

<unused8>

-0.71

POSITIVE LOGITS

 autoridad

0.33

return

0.32

 sozinho

0.32

 Begründung

0.32

 autorización

0.31

 uits

0.30

 odpowiedzi

0.30

 protesta

0.30

 identidad

0.29

 Erklärung

0.29

Activations Density 0.013%

code structure and syntax within programming or scripting contexts

documentation and code structure

No Comments

No Known Activations