INDEX

Explanations

the term "newly" as well as variations of it

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

ewire

-0.18

/Dk

-0.17

isContained

-0.15

é¡¿

-0.15

EMPLARY

-0.14

ÐµÐ»Ð¸Ñĩ

-0.14

íħĶ

-0.14

chyb

-0.14

herits

-0.14

matchCondition

-0.14

POSITIVE LOGITS

 mint

0.44

mint

0.33

wed

0.30

 Mint

0.28

 christ

0.25

 arrived

0.25

 formed

0.23

wed

0.23

-open

0.22

-re

0.20

Activations Density 0.015%

the term "newly" as well as variations of it

No Comments

No Known Activations