INDEX

Explanations

negative assessments of health and cleanliness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

erras

-0.18

ëĭĿ

-0.15

SENT

-0.15

dÃ¼l

-0.14

issor

-0.14

.logout

-0.13

jeme

-0.13

rief

-0.13

PasswordEncoder

-0.13

.dds

-0.13

POSITIVE LOGITS

fil

0.40

 dirty

0.34

fil

0.34

 filthy

0.34

 rats

0.33

 mold

0.32

 filt

0.32

Fil

0.31

filt

0.30

Fil

0.29

Activations Density 0.166%

negative assessments of health and cleanliness

No Comments

No Known Activations