INDEX

Explanations

references to infidelity and marital issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

antity

-0.16

clusive

-0.15

 homophobic

-0.15

acci

-0.15

ylan

-0.15

 Einsatz

-0.15

essim

-0.14

opr

-0.14

#+#

-0.14

 persec

-0.14

POSITIVE LOGITS

 affairs

0.33

 affair

0.32

 Affairs

0.27

 dall

0.26

 poly

0.25

 extr

0.25

 phil

0.25

 serial

0.22

 bedding

0.22

 prom

0.22

Activations Density 0.247%

references to infidelity and marital issues

No Comments

No Known Activations