INDEX

Explanations

names or references related to a specific person or identity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ingt

-0.17

al

-0.16

ispers

-0.16

èĻŁ

-0.15

aksi

-0.15

 cÃ¡nh

-0.15

iasi

-0.15

alsy

-0.15

POSITIVE LOGITS

card

0.27

ci

0.24

hest

0.23

ardo

0.23

cards

0.21

igli

0.20

cio

0.20

eland

0.18

cit

0.18

Åĵur

0.18

Activations Density 0.013%

names or references related to a specific person or identity

No Comments

No Known Activations