INDEX

Explanations

terms associated with protection and damage prevention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

Affected

-0.14

 winners

-0.14

 victims

-0.14

 INTERRU

-0.14

engu

-0.13

elin

-0.13

undef

-0.13

ixels

-0.13

cef

-0.13

umar

-0.13

POSITIVE LOGITS

 harm

0.24

 harmful

0.23

 harms

0.22

 Harm

0.22

 potential

0.22

 external

0.21

 dangers

0.20

æĿ¥èĩª

0.20

 possible

0.20

 dele

0.19

Activations Density 0.127%

terms associated with protection and damage prevention

No Comments

No Known Activations