INDEX

Explanations

phrases indicating directional movement or transformation

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

 undermin

-0.17

ì¡±

-0.17

ened

-0.16

iller

-0.16

udder

-0.16

wyn

-0.15

ccb

-0.15

lah

-0.14

 jadx

-0.14

ries

-0.14

POSITIVE LOGITS

/out

0.36

/from

0.24

 obliv

0.22

 into

0.21

the

0.20

 adulthood

0.19

 Into

0.19

 existence

0.19

 action

0.18

Activations Density 0.120%

phrases indicating directional movement or transformation

No Comments

No Known Activations