INDEX

Explanations

terms related to weakness 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 @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

eer

-0.18

ione

-0.17

asca

-0.17

ing

-0.17

ionic

-0.16

ingham

-0.16

ional

-0.16

eu

-0.15

ION

-0.15

 Speedway

-0.15

POSITIVE LOGITS

-strong

0.28

å¼±

0.27

weak

0.26

 Weak

0.25

 weak

0.25

Weak

0.25

lings

0.23

ens

0.23

 weaker

0.23

ly

0.23

Activations Density 0.011%

terms related to weakness or vulnerability

No Comments

No Known Activations