INDEX

Explanations

elements of structured data, particularly in markup or programming contexts

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

oftware

-0.15

ãĤ¤ãĤº

-0.15

<typeof

-0.15

ç±į

-0.15

_mE

-0.14

Ã¤nde

-0.13

 Ð¾Ð´Ð¸

-0.13

iele

-0.13

 bows

-0.13

opak

-0.13

POSITIVE LOGITS

0.29

yes

0.28

0.26

Yes

0.25

yes

0.24

YES

0.24

 zero

0.23

 TRUE

0.22

Yes

0.21

YES

0.21

Activations Density 0.068%

elements of structured data, particularly in markup or programming contexts

No Comments

No Known Activations