INDEX

Explanations

elements related to collaboration and group activities

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

postalcode

-0.16

PGA

-0.16

cap

-0.15

orate

-0.15

odem

-0.14

vrd

-0.14

 Braun

-0.14

atalog

-0.13

akk

-0.13

åĩºçīĪç¤¾

-0.13

POSITIVE LOGITS

 performances

0.21

 night

0.20

 performers

0.20

 dancing

0.20

 event

0.20

_PI

0.19

 dances

0.18

 dance

0.18

night

0.18

 performance

0.17

Activations Density 0.002%

elements related to collaboration and group activities

No Comments

No Known Activations