INDEX

Explanations

references to USA Today and its variations or mentions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

eyh

-0.17

.Guna

-0.15

ilet

-0.15

ìĥĪ

-0.15

ledge

-0.15

ond

-0.15

ãģ£ãģı

-0.14

ause

-0.14

lege

-0.14

eken

-0.14

POSITIVE LOGITS

 TODAY

0.25

merican

0.20

 Today

0.19

.gov

0.18

gent

0.17

ID

0.17

opoly

0.17

 latter

0.17

meric

0.16

gain

0.15

Activations Density 0.020%

references to USA Today and its variations or mentions

No Comments

No Known Activations