INDEX

Explanations

references to screens and their features

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

cul

-0.20

riad

-0.17

rians

-0.17

extr

-0.16

utor

-0.15

epad

-0.15

ÑģÑı

-0.15

ÑģÑĸÐ¼

-0.15

ft

-0.15

ans

-0.15

POSITIVE LOGITS

å¹ķ

0.34

aver

0.23

avers

0.22

plays

0.22

writer

0.22

hots

0.22

writers

0.21

writing

0.20

 saver

0.19

.blit

0.19

Activations Density 0.023%

references to screens and their features

No Comments

No Known Activations