INDEX

Explanations

references to databases and research publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

lak

-0.15

å®ļ

-0.15

 Scratch

-0.14

ymax

-0.14

eos

-0.14

Tomorrow

-0.14

vider

-0.14

 Ð½Ð°Ð½ÐµÑģ

-0.14

ellation

-0.13

isoft

-0.13

POSITIVE LOGITS

 Worm

0.16

hev

0.14

ts

0.14

tones

0.14

 Ð²Ð¾ÑĤ

0.14

amin

0.13

 referencing

0.13

mobx

0.13

ModelProperty

0.13

zilla

0.13

Activations Density 0.004%

references to databases and research publications

No Comments

No Known Activations