INDEX

Explanations

references to academic and professional titles

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

utschen

-0.20

 deutschen

-0.20

olit

-0.19

ussen

-0.18

isten

-0.18

Ð½ÑİÑİ

-0.18

 eigenen

-0.17

anten

-0.16

 ÑįÑĤÑĥ

-0.16

 kleinen

-0.16

POSITIVE LOGITS

iger

0.30

licher

0.28

ischer

0.28

 aktu

0.27

ender

0.25

erner

0.24

ter

0.23

aler

0.23

abler

0.23

riger

0.22

Activations Density 0.027%

references to academic and professional titles

No Comments

No Known Activations