INDEX

Explanations

domain names and URLs in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

iaux

-0.19

ecko

-0.16

ignKey

-0.15

.CopyTo

-0.15

ALLERY

-0.15

cest

-0.15

.Sdk

-0.15

çłĤ

-0.14

pais

-0.14

tring

-0.14

POSITIVE LOGITS

zÃ¡

0.18

ode

0.17

indrome

0.15

WC

0.15

OME

0.14

ureka

0.14

WC

0.14

RG

0.14

endl

0.14

Activations Density 0.003%

domain names and URLs in the text

No Comments

No Known Activations