INDEX

Explanations

references to amusement parks and related activities

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

 Disney

-0.17

 Disneyland

-0.16

ingham

-0.16

Disney

-0.15

_dl

-0.15

ivet

-0.15

 Walt

-0.14

anche

-0.14

éĽħ

-0.14

pod

-0.14

POSITIVE LOGITS

 Universal

0.31

Universal

0.29

 UNIVERS

0.25

 universal

0.24

universal

0.22

 Univers

0.19

 univers

0.17

ë²Ķ

0.15

 Islands

0.15

 ThemeData

0.15

Activations Density 0.014%

references to amusement parks and related activities

No Comments

No Known Activations