INDEX

Explanations

incidents involving emergency responses and acts of heroism

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

 Humb

-0.15

ennon

-0.14

eyer

-0.14

à¥įà¤°à¤¸

-0.14

 Sender

-0.14

.cgi

-0.14

è°Ī

-0.14

_deinit

-0.13

mey

-0.13

_counters

-0.13

POSITIVE LOGITS

 helping

0.24

 help

0.23

 Helping

0.22

 intervention

0.21

 rescue

0.21

 helps

0.20

 Help

0.20

 resc

0.20

-help

0.20

Help

0.20

Activations Density 0.141%

incidents involving emergency responses and acts of heroism

No Comments

No Known Activations