INDEX

Explanations

specific historical events or milestones related to architecture or notable figures in the field

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

-0.18

Û±Û¹Û¸

-0.17

Û±Û¹Û¹

-0.17

 videot

-0.16

 vidÃ©os

-0.16

ergy

-0.16

 Laser

-0.16

fax

-0.15

 cloning

-0.15

POSITIVE LOGITS

0.37

0.33

0.32

0.29

0.27

0.23

-Nazi

0.21

 Depression

0.21

 WWII

0.20

WW

0.20

Activations Density 1.152%

specific historical events or milestones related to architecture or notable figures in the field

No Comments

No Known Activations