INDEX

Explanations

phrases indicating educational or reference value

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

rish

-0.15

èĢ

-0.15

.dtd

-0.14

Joined

-0.14

alie

-0.13

.mac

-0.13

ialect

-0.13

osu

-0.13

fixtures

-0.13

zw

-0.13

POSITIVE LOGITS

 anyone

0.34

 anybody

0.29

 Anyone

0.27

Anyone

0.26

 readers

0.24

 students

0.23

 audiences

0.23

lay

0.20

 professionals

0.20

 reader

0.20

Activations Density 0.156%

phrases indicating educational or reference value

No Comments

No Known Activations