INDEX

Explanations

references to individuals, particularly those with 'Sel' in their names, suggesting a focus on notable personalities or characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

dre

-0.17

rikes

-0.15

é¥

-0.14

odate

-0.14

fall

-0.13

illard

-0.13

obao

-0.13

ÃŃv

-0.13

ilater

-0.13

ìĦł

-0.13

POSITIVE LOGITS

inux

0.24

bst

0.23

Ã§uk

0.22

vester

0.21

wyn

0.20

engkap

0.19

ÐµÐºÑĤÐ¸Ð²

0.19

BST

0.18

anik

0.18

ction

0.18

Activations Density 0.007%

references to individuals, particularly those with 'Sel' in their names, suggesting a focus on notable personalities or characters

No Comments

No Known Activations