INDEX

Explanations

references to museums and historical regions

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

 Parkway

-0.17

mina

-0.15

è©

-0.14

acts

-0.14

lett

-0.14

ufs

-0.14

efa

-0.14

laz

-0.14

 ì°¸

-0.14

illet

-0.13

POSITIVE LOGITS

obl

0.16

 depos

0.15

Kra

0.15

ersen

0.15

æ¸©

0.15

ahlen

0.14

Rai

0.14

ÑģÑĤÑĮ

0.14

 Pref

0.13

Activations Density 0.044%

references to museums and historical regions

No Comments

No Known Activations