INDEX

Explanations

significant numerical values and dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

Bod

-0.15

pis

-0.14

ä¹İ

-0.14

hoe

-0.14

aight

-0.14

rais

-0.14

istrovstvÃŃ

-0.14

iph

-0.14

 Lilly

-0.14

erner

-0.13

POSITIVE LOGITS

avra

0.15

Ø§Ø¨ÙĦ

0.15

.addButton

0.15

andy

0.14

uez

0.14

WS

0.14

IFO

0.14

 à¸¡à¸«

0.13

 rendre

0.13

 lord

0.13

Activations Density 0.003%

significant numerical values and dates

No Comments

No Known Activations