INDEX

Explanations

code snippets related to user management and account configurations

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

content builder

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

 surla

-0.63

parsedMessage

-0.56

awtextra

-0.56

verwijspagina

-0.56

ⓧ

-0.55

 defaultstate

-0.54

évaluateur

-0.54

adaptiveStyles

-0.53

DockStyle

-0.53

Билгалдахарш

-0.52

POSITIVE LOGITS

ListModule

0.44

 program

0.42

 module

0.40

élien

0.39

щита

0.35

 Department

0.35

 department

0.35

 Identity

0.35

 departments

0.34

PM

0.34

Activations Density 0.071%

code snippets related to user management and account configurations

content builder

No Comments

No Known Activations