INDEX

Explanations

different forms and uses of the word "record" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

imoto

-0.18

akh

-0.17

idor

-0.17

utom

-0.16

isi

-0.15

wel

-0.15

endor

-0.14

okt

-0.14

MLElement

-0.14

ugs

-0.14

POSITIVE LOGITS

-breaking

0.31

keeping

0.29

breaking

0.23

-setting

0.19

atorio

0.19

ant

0.18

ings

0.17

set

0.17

 breaking

0.17

ammers

0.17

Activations Density 0.029%

different forms and uses of the word "record" in various contexts

No Comments

No Known Activations