INDEX

Explanations

references to physical vulnerabilities or safety equipment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

airo

-0.16

à¤ļà¤¨

-0.15

zych

-0.14

 Ø¨ÙĨØ¯ÛĮ

-0.14

oen

-0.14

esh

-0.14

omics

-0.14

ady

-0.14

anke

-0.14

 overd

-0.14

POSITIVE LOGITS

 launcher

0.17

è£ħç½®

0.16

 Newman

0.16

 cleaner

0.16

-type

0.15

 apparatus

0.15

 blocker

0.14

 Ware

0.14

alyzer

0.14

-style

0.14

Activations Density 0.278%

references to physical vulnerabilities or safety equipment

No Comments

No Known Activations