INDEX

Explanations

phrases and context related to authoritative statements or proclamations

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

 Shiite

-0.25

 Shepard

-0.22

fony

-0.22

 LGBTQ

-0.21

 whiskey

-0.20

 Francois

-0.20

errick

-0.20

 WiFi

-0.19

 advisors

-0.19

 Griffin

-0.18

POSITIVE LOGITS

 Muhammed

0.29

 Andersen

0.28

 MacDonald

0.27

 Nicholson

0.25

 Henderson

0.25

0.24

 Patterson

0.23

 WIFI

0.23

ensen

0.22

 Mohammed

0.22

Activations Density 0.181%

phrases and context related to authoritative statements or proclamations

No Comments

No Known Activations