INDEX

Explanations

tour-related phrases and experiences

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

adge

-0.15

istar

-0.15

(æľĪ

-0.14

 Speak

-0.14

Ã¤ge

-0.14

 Exclude

-0.13

 Seek

-0.13

ulp

-0.13

 adher

-0.13

gne

-0.13

POSITIVE LOGITS

 experience

0.24

 enjoy

0.23

get

0.22

 Experience

0.22

 become

0.21

Experience

0.21

experience

0.19

Get

0.19

 discover

0.18

Enjoy

0.18

Activations Density 0.299%

tour-related phrases and experiences

No Comments

No Known Activations