INDEX

Explanations

instances of loss and victimization involving individuals or groups

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

Kill

-0.23

 Kill

-0.21

 kill

-0.21

kill

-0.19

 killing

-0.18

 Kills

-0.18

 kills

-0.16

 Killing

-0.15

kills

-0.14

 bÃło

-0.14

POSITIVE LOGITS

 died

0.27

die

0.26

 succ

0.26

DIE

0.24

 dies

0.23

 perish

0.23

die

0.23

Die

0.21

 suff

0.21

_die

0.20

Activations Density 0.100%

instances of loss and victimization involving individuals or groups

No Comments

No Known Activations