INDEX

Explanations

references to educational institutions and academic content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

quier

-0.16

achment

-0.16

-0.15

deÅŁ

-0.15

è¹

-0.15

rych

-0.15

romium

-0.14

ĶåĽŀ

-0.14

.slot

-0.14

Ð¾Ð»Ñİ

-0.14

POSITIVE LOGITS

Bil

0.28

 Gran

0.27

Pam

0.25

Zar

0.25

Cad

0.25

Ov

0.24

Gran

0.24

Mad

0.24

 Burg

0.23

 Vall

0.22

Activations Density 0.023%

references to educational institutions and academic content

No Comments

No Known Activations

references to educational institutions and academic content

No Comments

No Known Activations