INDEX

Explanations

references to pendulums and related terms

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

smith

-0.18

utherford

-0.18

å¯¸

-0.15

erot

-0.15

à¸±à¸Ļà¸ķ

-0.15

TEL

-0.15

eru

-0.15

erÃ§

-0.14

erule

-0.14

éĨ

-0.14

POSITIVE LOGITS

ulum

0.29

leton

0.27

 pend

0.26

 Pend

0.22

ennis

0.20

pend

0.19

ulous

0.19

PEND

0.17

ragon

0.17

 hanging

0.16

Activations Density 0.006%

references to pendulums and related terms

No Comments

No Known Activations