INDEX

Explanations

instances of the word "analysis."

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

lane

-0.16

ening

-0.16

etail

-0.16

/sl

-0.15

Ã¼zel

-0.15

nant

-0.15

heits

-0.15

sworth

-0.15

loe

-0.15

ality

-0.14

POSITIVE LOGITS

ogue

0.22

yses

0.21

tical

0.21

/design

0.19

Ú¯Ø±Ø§ÙĨ

0.18

 paralysis

0.17

conda

0.16

ogs

0.16

uger

0.16

ges

0.15

Activations Density 0.031%

instances of the word "analysis."

No Known Activations