INDEX

Explanations

interrogative words or phrases indicating questions

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

azen

-0.18

ieval

-0.16

ecta

-0.15

lor

-0.15

ideon

-0.15

ured

-0.14

jejer

-0.14

esting

-0.14

esti

-0.14

à¸¸à¸Ļ

-0.14

POSITIVE LOGITS

enever

0.28

ilst

0.24

ose

0.23

ether

0.22

isper

0.22

ETHER

0.22

eras

0.21

ichever

0.20

atever

0.20

EN

0.20

Activations Density 0.009%

interrogative words or phrases indicating questions

No Comments

No Known Activations