INDEX

Explanations

words related to movement and transitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

ark

-0.16

nder

-0.15

uilt

-0.15

ilder

-0.14

bas

-0.14

atas

-0.14

alo

-0.14

kem

-0.14

alb

-0.14

 propos

-0.14

POSITIVE LOGITS

 into

0.32

 onto

0.30

into

0.26

onto

0.22

 Into

0.21

 back

0.21

 naar

0.20

 vÃło

0.19

à¹Ħà¸Ľà¸¢

0.19

Into

0.19

Activations Density 0.167%

words related to movement and transitions

No Comments

No Known Activations