INDEX

Explanations

references to specific individuals, particularly names starting with the letter 'D'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

æ¡Ĳ

-0.18

avis

-0.17

ÑĢÑĥÐ³

-0.16

iaz

-0.16

ouble

-0.15

 cour

-0.15

uty

-0.15

£i

-0.15

emons

-0.14

Ú¯Ø§Ø±

-0.14

POSITIVE LOGITS

istrovstvÃŃ

0.19

opal

0.17

antan

0.16

nou

0.15

anel

0.15

allon

0.15

SB

0.15

(strict

0.14

ktop

0.14

 blond

0.14

Activations Density 0.046%

references to specific individuals, particularly names starting with the letter 'D'

No Comments

No Known Activations