INDEX

Explanations

references to well-known celebrities or public figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ofs

-0.16

akit

-0.14

RVA

-0.14

otton

-0.14

bras

-0.13

BCM

-0.13

 cocci

-0.13

elves

-0.13

/Foundation

-0.13

 kraje

-0.13

POSITIVE LOGITS

 Fucking

0.20

mania

0.18

 Freak

0.18

 Netanyahu

0.17

 freak

0.17

’s

0.17

vs

0.17

's

0.17

-esque

0.17

 Gandhi

0.16

Activations Density 0.101%

references to well-known celebrities or public figures

No Comments

No Known Activations