INDEX

Explanations

terms related to web security and data protection

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

 Unsupported

-0.15

 porr

-0.15

unsupported

-0.14

 contraction

-0.14

 hatch

-0.14

 contracted

-0.14

tti

-0.14

IMER

-0.13

exter

-0.13

ags

-0.13

POSITIVE LOGITS

 attacker

0.28

 attackers

0.25

MIT

0.24

 adversaries

0.23

Mit

0.23

 Evil

0.23

Eve

0.22

 attacks

0.22

 attack

0.22

Forge

0.22

Activations Density 0.058%

terms related to web security and data protection

No Comments

No Known Activations