INDEX

Explanations

the word "similar" and its variations, indicating comparisons or likenesses

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

efeller

-0.19

een

-0.17

rac

-0.16

uyen

-0.15

ses

-0.15

vn

-0.15

vs

-0.14

ey

-0.14

eenth

-0.14

Vien

-0.14

POSITIVE LOGITS

ily

0.45

-minded

0.29

-sized

0.28

ities

0.28

-aged

0.27

 minded

0.27

-sex

0.26

 sized

0.26

 vein

0.22

-looking

0.22

Activations Density 0.033%

the word "similar" and its variations, indicating comparisons or likenesses

No Comments

No Known Activations