INDEX

Explanations

references to violence and medical procedures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

iral

-0.17

tring

-0.16

arine

-0.15

abi

-0.15

DEM

-0.15

 ApiController

-0.15

ieg

-0.14

Î±Î»

-0.14

 Stanton

-0.14

erie

-0.14

POSITIVE LOGITS

 needle

0.28

needle

0.26

 needles

0.25

stab

0.23

 Needle

0.22

æıĴ

0.20

 stabbed

0.18

 stabbing

0.18

 inserted

0.18

 insertion

0.17

Activations Density 0.089%

references to violence and medical procedures

No Comments

No Known Activations