INDEX

Explanations

statements regarding the status or identity of places or regions

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

atten

-0.17

omba

-0.15

Ð±Ð¸ÑĤ

-0.15

ullen

-0.14

ancel

-0.14

eba

-0.14

inea

-0.14

 Ã¶ÄŁ

-0.14

amus

-0.13

 ëķ

-0.13

POSITIVE LOGITS

 home

0.41

home

0.27

-home

0.26

 host

0.26

 Home

0.24

Home

0.23

(home

0.22

 sede

0.21

 HOME

0.21

 known

0.21

Activations Density 0.114%

statements regarding the status or identity of places or regions

No Comments

No Known Activations