INDEX

Explanations

geographical locations and landmarks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 airports

-0.20

 Resorts

-0.19

Airport

-0.18

 CITY

-0.18

 City

-0.16

 city

-0.16

 cities

-0.16

 Harbour

-0.16

(city

-0.16

 Strait

-0.16

POSITIVE LOGITS

 neighborhood

0.21

 district

0.19

 neigh

0.18

:"-"`↵

0.17

-area

0.17

 Square

0.17

 station

0.17

 area

0.17

wyn

0.17

 branch

0.16

Activations Density 0.154%

geographical locations and landmarks

No Comments

No Known Activations