INDEX

Explanations

references to scientific research articles and publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

/Linux

-0.19

atsu

-0.17

 lifelong

-0.15

 nÄĥ

-0.15

aw

-0.15

/to

-0.15

ë¬¼

-0.15

yonel

-0.15

steen

-0.14

leton

-0.14

POSITIVE LOGITS

icrous

0.22

itud

0.20

iferay

0.17

forms

0.17

acies

0.17

ifestyles

0.17

ardo

0.17

itudes

0.15

quo

0.15

Ð¸ÑĢÑĥ

0.15

Activations Density 1.577%

references to scientific research articles and publications

No Comments

No Known Activations