INDEX

Explanations

positive descriptors related to content and resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

á»ı

-0.16

 Derrick

-0.15

/trunk

-0.14

nan

-0.14

 conspir

-0.14

chop

-0.14

cos

-0.13

 tied

-0.13

stead

-0.13

ÃŃsto

-0.13

POSITIVE LOGITS

åħ¶ä¸Ń

0.26

 here

0.23

 à¤ĩà¤¸à¤®

0.21

 therein

0.21

 contained

0.21

here

0.20

 burada

0.20

contained

0.20

è¿ĻéĩĮ

0.19

_here

0.19

Activations Density 0.218%

positive descriptors related to content and resources

No Comments

No Known Activations

positive descriptors related to content and resources

No Comments

No Known Activations