INDEX

Explanations

references to themes and thematic elements in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

aneous

-0.18

arian

-0.18

teen

-0.17

ty

-0.17

inet

-0.16

iegel

-0.15

ree

-0.15

hti

-0.15

nd

-0.15

land

-0.14

POSITIVE LOGITS

elves

0.24

æĿĲ

0.21

atically

0.19

 park

0.19

0.18

atical

0.17

eted

0.16

gth

0.16

eting

0.16

Setter

0.16

Activations Density 0.017%

references to themes and thematic elements in the text

No Comments

No Known Activations