INDEX

Explanations

references to death or dying

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

weise

-0.18

.tt

-0.17

.GroupLayout

-0.15

bildung

-0.15

credible

-0.15

ial

-0.15

áº¡y

-0.15

adera

-0.14

desc

-0.14

mesinin

-0.14

POSITIVE LOGITS

 Dead

0.24

 dead

0.22

Dead

0.22

lier

0.22

.dead

0.21

ening

0.20

locks

0.19

 DEAD

0.19

ened

0.19

dead

0.19

Activations Density 0.018%

references to death or dying

No Comments

No Known Activations