INDEX

Explanations

questions and negative statements regarding societal issues or personal beliefs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

you

-0.27

you

-0.25

You

-0.25

You

-0.24

_you

-0.22

YOU

-0.21

YOU

-0.20

-you

-0.20

éĤ£äºĽ

-0.18

>You

-0.18

POSITIVE LOGITS

olated

0.26

chem

0.20

olate

0.20

abella

0.20

 there

0.20

olation

0.20

 anyone

0.20

 anybody

0.20

aac

0.19

nt

0.18

Activations Density 0.108%

questions and negative statements regarding societal issues or personal beliefs

No Comments

No Known Activations