INDEX

Explanations

references to pendulums and related concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

smith

-0.19

utherford

-0.17

herits

-0.17

ë§¥

-0.16

iefs

-0.15

altung

-0.14

å¯¸

-0.14

éĨ

-0.14

er

-0.14

/***/

-0.14

POSITIVE LOGITS

ulum

0.30

leton

0.28

 pend

0.24

 Pend

0.22

ennis

0.21

ulous

0.19

pend

0.18

PEND

0.18

ragon

0.18

Ore

0.16

Activations Density 0.007%

references to pendulums and related concepts

No Comments

No Known Activations