INDEX

Explanations

references to clicking links or buttons in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

quine

-0.16

.azure

-0.15

ãĥ¼ãĤ¹

-0.15

 Lists

-0.15

baj

-0.14

 pent

-0.14

lis

-0.14

 Poll

-0.14

 Trib

-0.13

uss

-0.13

POSITIVE LOGITS

ackle

0.16

 ÑıÐºÑĥ

0.15

ÑģÐ»Ð¸

0.14

 YYSTYPE

0.14

abcdefghijklmnop

0.14

æ´¥

0.14

slashes

0.13

dux

0.13

opis

0.13

eriod

0.13

Activations Density 0.016%

references to clicking links or buttons in the text

No Comments

No Known Activations