INDEX

Explanations

phrases related to performance and entertainment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 à¸¡à¸«

-0.16

 xlink

-0.15

ikit

-0.15

dice

-0.15

aurus

-0.15

]._

-0.14

aday

-0.14

succ

-0.14

isl

-0.14

ÐµÐ¹Ð½

-0.14

POSITIVE LOGITS

DW

0.31

DW

0.29

 Dancing

0.28

 dancing

0.27

 Mirror

0.26

 dance

0.26

 pros

0.25

 Strict

0.25

 dances

0.25

 ball

0.24

Activations Density 0.002%

phrases related to performance and entertainment

No Comments

No Known Activations