INDEX

Explanations

phrases that reference a group of people and their experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

å¿Ĺ

-0.15

alse

-0.15

 Ð°Ð¼

-0.15

Ø·

-0.14

å¤§åħ¨

-0.14

å®ľ

-0.14

äººäºº

-0.13

ree

-0.13

Exchange

-0.13

tub

-0.13

POSITIVE LOGITS

 unfamiliar

0.23

 wondering

0.23

 missed

0.18

 familiar

0.18

 wondered

0.17

 wonder

0.17

asca

0.17

Wonder

0.16

ovit

0.16

 eagle

0.16

Activations Density 0.034%

phrases that reference a group of people and their experiences

No Comments

No Known Activations