INDEX

Explanations

references to betting or gambling-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

itarian

-0.19

iero

-0.15

oles

-0.15

bine

-0.15

_tid

-0.14

hower

-0.14

ergic

-0.14

 åĶ

-0.14

iline

-0.14

Ã¤tz

-0.14

POSITIVE LOGITS

0.26

ters

0.26

tors

0.24

fair

0.24

rays

0.23

ray

0.23

WEEN

0.21

tings

0.21

tor

0.20

lehem

0.19

Activations Density 0.017%

references to betting or gambling-related terms

No Comments

No Known Activations

references to betting or gambling-related terms

No Comments

No Known Activations