INDEX

Explanations

references to magic and magical elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

PRNewswire

-0.38

논

-0.38

ình

-0.35

 arşivlendi

-0.34

 contextual

-0.34

 seizure

-0.34

 Мексичка

-0.33

 auseinander

-0.33

 Seizures

-0.32

vânt

-0.31

POSITIVE LOGITS

 magic

1.38

magic

1.20

Magic

1.19

 Magic

1.19

 MAGIC

1.12

MAGIC

1.05

 magical

1.05

 magia

1.04

 mág

1.00

 mágica

1.00

Activations Density 1.434%

references to magic and magical elements

No Comments

No Known Activations