INDEX

Explanations

references to legal processes and documented evidence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

á»Ľp

-0.17

asbourg

-0.15

hx

-0.15

 Monster

-0.14

 lÆ°Æ¡ng

-0.14

æĿŁ

-0.14

 porad

-0.14

/Dk

-0.14

akest

-0.14

 permalink

-0.14

POSITIVE LOGITS

dem

0.22

 replication

0.22

 bill

0.19

 suit

0.19

 assignments

0.19

 suits

0.18

 eject

0.18

tro

0.18

 venue

0.18

 traverse

0.18

Activations Density 0.152%

references to legal processes and documented evidence

No Comments

No Known Activations