INDEX

Explanations

concepts related to secrecy, trust, and deception in relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

roz

-0.16

ÑĢÑĥÐ±

-0.15

fol

-0.14

anut

-0.14

alarÄ±ndan

-0.14

ÑĢÐµÐ¼Ñı

-0.14

edla

-0.14

 inert

-0.14

irim

-0.13

 Serious

-0.13

POSITIVE LOGITS

 hiding

0.33

 secrets

0.32

 secrecy

0.30

 conceal

0.29

 secretive

0.28

 secret

0.28

éļł

0.27

 hides

0.26

 hidden

0.25

 hide

0.25

Activations Density 0.141%

concepts related to secrecy, trust, and deception in relationships

No Comments

No Known Activations