INDEX

Explanations

references to employees and employment-related terminology

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

employee

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

verwijspagina

-0.48

 difusión

-0.45

 acuarela

-0.44

 logística

-0.44

 pintada

-0.43

 pegatinas

-0.43

cocina

-0.43

 aventuras

-0.43

 aislada

-0.42

 dibujos

-0.42

POSITIVE LOGITS

 Employee

1.52

 employee

1.47

Employee

1.38

employee

1.34

 Employees

1.33

 employees

1.31

 employe

1.24

 EMPLOYEE

1.21

Employees

1.21

employees

1.20

Activations Density 0.018%

references to employees and employment-related terminology

employee

No Comments

No Known Activations