INDEX

Explanations

references to mythology and historical narratives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AllWindows

-0.16

Ã®tre

-0.15

glfw

-0.15

 Anglic

-0.15

ctrine

-0.15

æľ

-0.14

SystemService

-0.14

 Dickens

-0.14

ivr

-0.14

 à¤®à¤¸

-0.14

POSITIVE LOGITS

 Greek

0.46

Greek

0.39

 Greece

0.39

 Greeks

0.36

 Athens

0.33

 Athen

0.30

 Zeus

0.30

 Olymp

0.29

 myth

0.29

 gods

0.28

Activations Density 0.176%

references to mythology and historical narratives

No Comments

No Known Activations