INDEX

Explanations

expressions indicating comparison or addition of information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ismet

-0.18

anners

-0.15

iglia

-0.14

kur

-0.14

sen

-0.14

 trÃº

-0.14

anny

-0.14

 Encounter

-0.13

irting

-0.13

POSITIVE LOGITS

 than

0.34

-than

0.27

_than

0.24

 niÅ¼

0.24

 THAN

0.24

than

0.22

 Ã¤n

0.22

Than

0.20

 Than

0.20

á»Ļt

0.20

Activations Density 0.052%

expressions indicating comparison or addition of information

No Comments

No Known Activations