INDEX

Explanations

negations and negative phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

overy

-0.18

ayo

-0.16

uzzi

-0.15

uze

-0.14

ogens

-0.14

SSF

-0.13

loth

-0.13

etty

-0.13

ughter

-0.13

Quad

-0.13

POSITIVE LOGITS

 outlet

0.21

 sale

0.19

-sale

0.18

 cheap

0.18

 online

0.18

online

0.17

 Outlet

0.17

discount

0.17

 Sale

0.16

agli

0.16

Activations Density 0.033%

negations and negative phrases

No Comments

No Known Activations