INDEX

Explanations

references to motorsports and racing events

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

igne

-0.15

CONDITION

-0.15

_weak

-0.14

.nih

-0.14

 æ¥Ń

-0.14

erland

-0.14

ivor

-0.13

 meis

-0.13

alat

-0.13

lier

-0.13

POSITIVE LOGITS

 Indy

0.25

 Indianapolis

0.20

Parking

0.19

 parking

0.18

 Pens

0.17

 Buzz

0.17

Indian

0.16

 qualifying

0.16

 Indiana

0.16

 Lime

0.15

Activations Density 0.012%

references to motorsports and racing events

No Comments

No Known Activations