INDEX

Explanations

references to religious concepts and doctrines regarding morality and gender identity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

 humili

-0.16

textures

-0.15

cascade

-0.15

.Xtra

-0.15

leccion

-0.15

roz

-0.15

rames

-0.14

 homophobic

-0.14

presso

-0.14

Ð´Ð¸Ð½Ð°

-0.14

POSITIVE LOGITS

pag

0.23

 idol

0.21

 Nicol

0.21

 false

0.20

pag

0.20

 proud

0.19

 pride

0.19

 False

0.18

 pagan

0.18

 Idol

0.17

Activations Density 0.350%

references to religious concepts and doctrines regarding morality and gender identity

No Comments

No Known Activations