INDEX

Explanations

words related to the concept of "safety" or "security."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

erah

-0.15

slack

-0.15

atings

-0.15

aneously

-0.15

REA

-0.15

atee

-0.15

spy

-0.15

 refl

-0.14

askell

-0.14

setattr

-0.14

POSITIVE LOGITS

osten

0.28

ott

0.27

vil

0.27

osp

0.26

ottom

0.26

vol

0.26

otto

0.25

periment

0.25

ulla

0.25

ulle

0.25

Activations Density 0.006%

words related to the concept of "safety" or "security."

No Comments

No Known Activations