INDEX

Explanations

strongly emphasized phrases that convey absoluteness or completeness

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Datuak

-0.74

 Plated

-0.69

 רבה

-0.69

 Kontakte

-0.68

nameLabel

-0.68

 רבים

-0.67

Lieferumfang

-0.66

 Recife

-0.65

 Constellation

-0.65

 Miquel

-0.65

POSITIVE LOGITS

 totally

1.46

 completely

1.45

 Completely

1.43

 Totally

1.40

Totally

1.39

Completely

1.35

totally

1.26

 entirely

1.24

completely

1.21

 utterly

1.11

Activations Density 0.126%

strongly emphasized phrases that convey absoluteness or completeness

No Comments

No Known Activations