INDEX

Explanations

elements related to structured data and programming logic

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ToOne

-0.17

Â¹

-0.16

imits

-0.16

HING

-0.15

âĤģ

-0.15

esch

-0.15

redi

-0.15

ingles

-0.14

Ð»Ð°Ð²Ð°

-0.14

Ã«n

-0.14

POSITIVE LOGITS

0.43

Û²

0.31

ï¼Ĵ

0.31

äºĮ

0.30

 second

0.28

 äºĮ

0.27

two

0.26

zwe

0.26

.two

0.26

âĤĤ

0.24

Activations Density 0.121%

elements related to structured data and programming logic

No Comments

No Known Activations