INDEX

Explanations

phrases related to copyright and content usage policies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

uchen

-0.19

ÑĸÑĪ

-0.16

 aware

-0.15

rÃ©

-0.14

 Hast

-0.14

klÃ¤

-0.14

 awareness

-0.14

 drafting

-0.13

-length

-0.13

POSITIVE LOGITS

ABEL

0.17

 Redistribution

0.16

hook

0.16

 attribution

0.15

(permission

0.15

alion

0.15

olik

0.14

 permission

0.14

acker

0.14

oppers

0.14

Activations Density 0.152%

phrases related to copyright and content usage policies

No Comments

No Known Activations