INDEX

Explanations

instances of placeholder text commonly used in design and typesetting, particularly "Lorem ipsum."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 {}).

-0.15

enson

-0.15

riere

-0.14

 ÐºÑĢÐ¸

-0.14

,:,

-0.14

licative

-0.14

 Falk

-0.13

odv

-0.13

agal

-0.13

otos

-0.13

POSITIVE LOGITS

enderit

0.21

 labor

0.20

 magna

0.20

 incididunt

0.19

etur

0.18

 repreh

0.18

 aliqua

0.17

 adipisicing

0.17

 anim

0.17

 volupt

0.17

Activations Density 0.020%

instances of placeholder text commonly used in design and typesetting, particularly "Lorem ipsum."

No Comments

No Known Activations