INDEX

Explanations

phrases and actions related to setting and burning flags

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ully

-0.17

elman

-0.16

'icon

-0.16

-0.15

 crowds

-0.15

Away

-0.14

 Wilkinson

-0.14

ALI

-0.14

qua

-0.14

alion

-0.14

POSITIVE LOGITS

af

0.38

 astr

0.34

ag

0.34

aw

0.31

ask

0.30

ast

0.28

ash

0.26

afl

0.26

astr

0.25

 aire

0.25

Activations Density 0.174%

phrases and actions related to setting and burning flags

No Comments

No Known Activations