INDEX

Explanations

numerical information or statistics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

arpa

-0.17

otine

-0.17

anson

-0.16

å¤ı

-0.15

Ã³s

-0.15

prung

-0.15

aug

-0.14

rox

-0.14

Ã¸d

-0.14

orton

-0.14

POSITIVE LOGITS

tober

0.27

October

0.25

-Nov

0.24

Oct

0.23

Oct

0.22

oct

0.22

oct

0.21

/oct

0.21

 October

0.20

ä»½

0.19

Activations Density 0.012%

numerical information or statistics

No Comments

No Known Activations