INDEX

Explanations

phrases expressing openness and accessibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 Opening

-0.20

 openings

-0.20

 opening

-0.20

opening

-0.19

Opening

-0.19

 opener

-0.18

Opens

-0.17

-opening

-0.16

zik

-0.16

ç¹Ķ

-0.15

POSITIVE LOGITS

 closed

0.25

Closed

0.24

 Closed

0.24

closed

0.23

 possibility

0.22

 possibilit

0.20

 possibilities

0.20

-ended

0.19

 CLOSED

0.19

 welcoming

0.18

Activations Density 0.073%

phrases expressing openness and accessibility

No Comments

No Known Activations

phrases expressing openness and accessibility

No Comments

No Known Activations