INDEX

Explanations

comparative phrases that highlight superiority or significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

conj

-0.14

atrice

-0.14

ghi

-0.14

chos

-0.14

à¤¾à¤Ĺà¤¤

-0.14

çĭ¼

-0.13

AGO

-0.13

usi

-0.13

ocy

-0.13

ola

-0.13

POSITIVE LOGITS

 anyone

0.38

 anybody

0.37

any

0.35

 anywhere

0.32

ä»»ä½ķ

0.28

 than

0.27

 Anyone

0.27

Anyone

0.26

any

0.26

 others

0.26

Activations Density 0.116%

comparative phrases that highlight superiority or significance

No Comments

No Known Activations