INDEX

Explanations

terms related to medical conditions and physiological processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ousse

-0.18

ÑĥÐ»Ð¸

-0.15

cratch

-0.15

 Aggregate

-0.15

REAK

-0.14

aurant

-0.14

andest

-0.14

tplib

-0.13

anela

-0.13

.mu

-0.13

POSITIVE LOGITS

 oxygen

0.45

 Oxygen

0.40

xygen

0.31

 breathing

0.30

 breathed

0.30

CO

0.29

 saturation

0.28

CO

0.28

gas

0.28

 carbon

0.27

Activations Density 0.034%

terms related to medical conditions and physiological processes

No Comments

No Known Activations