INDEX

Explanations

instances of direct statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

agos

-0.18

rzy

-0.15

<decltype

-0.15

 Å¾Ã¡dnÃ©

-0.14

axed

-0.13

orrh

-0.13

 Ð¾Ð½Ð¾

-0.13

éĤ£éĩĮ

-0.13

Äįin

-0.13

dyby

-0.13

POSITIVE LOGITS

 prior

0.28

 Prior

0.26

 Additionally

0.26

Prior

0.24

 additionally

0.23

Additionally

0.23

 recent

0.23

 currently

0.22

 Currently

0.22

 Among

0.21

Activations Density 0.097%

instances of direct statements

No Comments

No Known Activations