INDEX

Explanations

terms related to safety and emergency preparedness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

Wunused

-0.17

istrovstvÃŃ

-0.16

_probe

-0.15

cie

-0.15

olta

-0.14

çĵľ

-0.14

 Trab

-0.14

.unwrap

-0.14

Argument

-0.14

_stderr

-0.14

POSITIVE LOGITS

 evacuation

0.25

 emergency

0.24

 Emergency

0.22

0.19

 rescue

0.19

 evac

0.18

 evacuate

0.17

 surviv

0.17

 emergencies

0.17

emergency

0.17

Activations Density 0.167%

terms related to safety and emergency preparedness

No Comments

No Known Activations