INDEX

Explanations

terms associated with safety in cyber-physical systems

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ettel

-0.15

à¸¶à¸ģ

-0.15

CAST

-0.15

ichel

-0.14

icle

-0.14

fgang

-0.14

iosa

-0.14

_ASSUME

-0.14

ritz

-0.14

atown

-0.14

POSITIVE LOGITS

kovi

0.15

ledo

0.15

iras

0.15

ck

0.15

 design

0.14

ALLOW

0.14

uster

0.14

kl

0.14

tra

0.13

Dur

0.13

Activations Density 0.246%

terms associated with safety in cyber-physical systems

No Comments

No Known Activations