INDEX

Explanations

punctuation and formatting characters commonly used 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 @ 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

the

-0.18

 there

-0.18

 this

-0.16

 when

-0.16

it

-0.15

for

-0.15

in

-0.15

 theirs

-0.15

 yours

-0.15

POSITIVE LOGITS

allen

0.15

elmet

0.15

Ã©ric

0.14

 RectTransform

0.14

TERN

0.14

vise

0.13

 UITableViewController

0.13

enery

0.13

>NN

0.13

ìłł

0.13

Activations Density 0.040%

punctuation and formatting characters commonly used in written text

No Comments

No Known Activations