INDEX

Explanations

references to cultural or historical significance

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

onta

-0.16

 INTERRUPTION

-0.14

elter

-0.14

éĺµ

-0.14

ipe

-0.14

.dtd

-0.14

.adv

-0.14

ropped

-0.13

arness

-0.13

à¸µà¸ģ

-0.13

POSITIVE LOGITS

 evolution

0.27

 impact

0.26

 role

0.26

 lives

0.24

 ways

0.24

 relationship

0.24

 rise

0.23

 roles

0.23

impact

0.23

 effect

0.22

Activations Density 0.171%

references to cultural or historical significance

No Comments

No Known Activations