INDEX

Explanations

references to popular culture, specifically film and television

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

ukkan

-0.16

à¸¢à¸ĩ

-0.15

 categorical

-0.15

uted

-0.15

verture

-0.14

Jim

-0.14

asin

-0.14

zes

-0.14

Po

-0.14

 Moreno

-0.14

POSITIVE LOGITS

mac

0.30

(mac

0.30

Mac

0.28

Mac

0.28

.mac

0.26

/mac

0.25

mac

0.25

MAC

0.24

MAC

0.23

_mac

0.20

Activations Density 0.016%

references to popular culture, specifically film and television

No Comments

No Known Activations