INDEX

Explanations

words associated with happiness and positive emotions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ilenames

-0.15

ussen

-0.15

 Beaut

-0.15

 Erotik

-0.14

ermen

-0.14

PHY

-0.14

aname

-0.14

olls

-0.14

casts

-0.14

POSITIVE LOGITS

-go

0.34

 ending

0.25

 Ending

0.24

 camper

0.24

 endings

0.23

-ever

0.21

-medium

0.21

-ending

0.21

Ending

0.20

/content

0.19

Activations Density 0.027%

words associated with happiness and positive emotions

No Comments

No Known Activations