INDEX

Explanations

references to corporate titles and leadership roles

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

New Auto-Interp

Configuration

fnlp/Llama3_1-8B-Base-LXM-8x/Llama3_1-8B-Base-L22M-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_mlp_out

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

ibel

-0.18

ILA

-0.16

.ev

-0.15

iras

-0.15

mor

-0.14

itioner

-0.14

xon

-0.14

anel

-0.13

zsche

-0.13

vester

-0.13

POSITIVE LOGITS

 Hubb

0.17

 Huck

0.15

pNet

0.15

Ð¾ÑĩÐ½Ð¾

0.15

pling

0.14

 Bridges

0.14

uggle

0.14

 Carter

0.14

Ã¡ce

0.14

upply

0.14

Activations Density 0.071%

references to corporate titles and leadership roles

No Comments

No Known Activations