INDEX

Explanations

elements related to academic publication metadata

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

UN

-0.16

pond

-0.15

([]*

-0.15

 Tomorrow

-0.15

 Watts

-0.15

Jur

-0.14

 Manuals

-0.14

#region

-0.14

Ķ

-0.14

asta

-0.14

POSITIVE LOGITS

 ACCEPT

0.23

 Accepted

0.23

.Accept

0.23

 accepted

0.22

 acceptance

0.21

Accepted

0.21

 acept

0.20

oola

0.20

accepted

0.20

 peer

0.19

Activations Density 0.013%

elements related to academic publication metadata

No Comments

No Known Activations