INDEX

Explanations

references to honors, awards, and achievements

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

 Ø¯Ø¹

-0.17

Ð»Ð¾Ðº

-0.15

zad

-0.15

ulis

-0.15

ernote

-0.15

ober

-0.14

/GPL

-0.14

ÑīÐ¸Ð½

-0.14

inki

-0.14

adÄ±k

-0.14

POSITIVE LOGITS

 Order

0.28

Order

0.24

 Orders

0.23

 decoration

0.23

 decorations

0.22

 order

0.22

 Decoration

0.22

 ORDER

0.20

-order

0.20

.Order

0.20

Activations Density 0.082%

references to honors, awards, and achievements

No Comments

No Known Activations