INDEX

Explanations

references to the concept of play

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

adx

-0.18

eman

-0.18

invert

-0.16

erot

-0.15

ska

-0.15

.dll

-0.15

-0.14

iales

-0.14

Ø¯ÛĮÚ¯Ø±

-0.14

erap

-0.14

POSITIVE LOGITS

ground

0.27

grounds

0.26

thing

0.26

fulness

0.26

acting

0.24

group

0.24

Ground

0.23

room

0.23

things

0.23

date

0.23

Activations Density 0.013%

references to the concept of play

No Comments

No Known Activations