INDEX

Explanations

items or components related to structured lists or items in a systematic format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

abay

-0.15

leftright

-0.15

abox

-0.14

Č↵

-0.14

ordum

-0.14

SGlobal

-0.13

Ø§ØŃÛĮ

-0.13

ãģıãĤī

-0.13

SAFE

-0.13

 ÑģÐºÐ°Ð·

-0.13

POSITIVE LOGITS

Å¡ek

0.16

 boom

0.15

erval

0.14

ector

0.14

iros

0.14

ectors

0.14

%x

0.14

antz

0.14

Ð¼Ð¸ÑĢ

0.14

soever

0.13

Activations Density 0.022%

items or components related to structured lists or items in a systematic format

No Comments

No Known Activations