INDEX

Explanations

incidents involving police and security interactions

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

antar

-0.17

uhan

-0.16

 SERIAL

-0.15

astreet

-0.15

(=)

-0.14

dete

-0.14

 decltype

-0.14

holm

-0.14

$MESS

-0.14

à¤¹à¤²

-0.14

POSITIVE LOGITS

 suff

0.16

lier

0.15

å±ħ

0.15

asse

0.14

Sad

0.14

ald

0.14

 instead

0.14

 then

0.14

 line

0.14

Activations Density 0.060%

incidents involving police and security interactions

No Comments

No Known Activations