INDEX

Explanations

phrases describing similarity or shared interests among individuals or ideas

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

eci

-0.15

¹Ħ

-0.14

ÙĬØ¬

-0.14

otty

-0.14

lash

-0.14

iste

-0.14

ogy

-0.14

à¤¶à¤¨

-0.14

ebi

-0.13

locate

-0.13

POSITIVE LOGITS

-minded

0.42

 minded

0.40

 Minds

0.28

 minds

0.28

able

0.26

hood

0.23

WISE

0.21

ably

0.21

abled

0.20

/dis

0.20

Activations Density 0.014%

phrases describing similarity or shared interests among individuals or ideas

No Known Activations