INDEX

Explanations

references to news organizations and broadcasting details

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

argout

-0.17

ä¸Ŀ

-0.16

nik

-0.15

Ook

-0.15

aller

-0.15

 Nickel

-0.14

Obl

-0.14

 nickel

-0.14

 Buffered

-0.13

tub

-0.13

POSITIVE LOGITS

ABC

0.41

ABC

0.37

abc

0.31

abc

0.30

_ABC

0.27

Disney

0.22

 Disney

0.21

 Walt

0.19

 ESPN

0.17

XYZ

0.17

Activations Density 0.029%

references to news organizations and broadcasting details

No Comments

No Known Activations