INDEX

Explanations

punctuation and structure in written text

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

fav

-0.16

yas

-0.15

lh

-0.15

Ð¿Ð¸

-0.15

zik

-0.15

Archive

-0.14

.WinForms

-0.14

 Sciences

-0.14

isters

-0.13

POSITIVE LOGITS

umba

0.15

ERIC

0.14

 itemprop

0.14

 catalog

0.14

umer

0.14

 Catalog

0.14

uide

0.14

oman

0.14

omen

0.13

AGO

0.13

Activations Density 0.003%

punctuation and structure in written text

No Comments

No Known Activations