INDEX

Explanations

references to art exhibitions and artistic works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

obel

-0.16

iegel

-0.15

 ApplicationException

-0.15

Å¡i

-0.15

ormap

-0.15

.cx

-0.14

icie

-0.14

Russ

-0.13

xsi

-0.13

endor

-0.13

POSITIVE LOGITS

 Vital

0.33

Gle

0.27

Lev

0.25

Fed

0.23

 Alexander

0.23

Ev

0.23

 Roman

0.21

Py

0.20

 Alex

0.20

 Maxim

0.19

Activations Density 0.071%

references to art exhibitions and artistic works

No Comments

No Known Activations