INDEX

Explanations

references to hacking and hacker-related terms

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

ores

-0.15

-0.14

toDouble

-0.14

romo

-0.14

odi

-0.13

agedList

-0.13

å¼

-0.13

vas

-0.13

POSITIVE LOGITS

ity

0.17

manship

0.15

tiv

0.14

ìłĲ

0.14

 Tham

0.14

esson

0.14

statt

0.14

ÑĤÐ¸Ð²

0.14

Activations Density 0.006%

references to hacking and hacker-related terms

No Comments

No Known Activations