INDEX

Explanations

references to client-related interactions and roles in a professional context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

 Wirksamkeit

-0.48

 Arbeiten

-0.43

 toepassing

-0.39

 niebla

-0.39

 bandeira

-0.37

 automatiques

-0.36

 voldo

-0.35

 Behör

-0.35

 referrerpolicy

-0.35

 oficjal

-0.35

POSITIVE LOGITS

 managers

1.07

 Sales

1.06

 sales

1.04

 marketing

1.00

 Marketing

0.98

 manager

0.95

Sales

0.94

 finance

0.93

HR

0.93

 Managers

0.93

Activations Density 0.415%

references to client-related interactions and roles in a professional context

No Comments

No Known Activations