INDEX

Explanations

phrases that express excessiveness or negativity in a situation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

abus

-0.14

rieb

-0.14

anness

-0.14

lier

-0.14

Ã¡vÄĽ

-0.14

à¸ļà¸£à¸£

-0.13

odus

-0.13

 olab

-0.13

æĽ´

-0.13

orthand

-0.13

POSITIVE LOGITS

 much

0.33

 Much

0.29

much

0.28

Much

0.26

 MUCH

0.24

 many

0.20

far

0.19

_MANY

0.19

 soon

0.18

many

0.18

Activations Density 0.031%

phrases that express excessiveness or negativity in a situation

No Comments

No Known Activations