INDEX

Explanations

phrases indicating risk or vulnerability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

uart

-0.15

mares

-0.15

.singleton

-0.14

 å¼ķ

-0.14

rove

-0.14

aires

-0.14

.Errors

-0.14

acity

-0.14

.prevent

-0.13

upal

-0.13

POSITIVE LOGITS

 increased

0.34

 heightened

0.28

 greater

0.27

 greatest

0.26

 Increased

0.25

greater

0.22

Increased

0.22

 imminent

0.20

 Greatest

0.20

Greater

0.19

Activations Density 0.017%

phrases indicating risk or vulnerability

No Comments

No Known Activations