INDEX

Explanations

future intentions or predictions expressed through the use of the word "will."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ancellable

-0.16

ectl

-0.15

edu

-0.14

oidal

-0.14

illions

-0.14

à¸Ńà¸²à¸Ī

-0.14

opr

-0.14

well

-0.14

istas

-0.14

quelle

-0.14

POSITIVE LOGITS

iams

0.37

iam

0.36

0.31

be

0.30

IAM

0.29

fully

0.29

kommen

0.28

 likely

0.26

ows

0.26

power

0.25

Activations Density 0.322%

future intentions or predictions expressed through the use of the word "will."

No Comments

No Known Activations