INDEX

Explanations

emotional and positive expressions related to experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

_by

-0.15

bes

-0.15

inae

-0.14

Which

-0.14

inis

-0.14

ByName

-0.14

orsi

-0.13

uby

-0.13

WithContext

-0.13

ãģ«ãĤĪãĤĭ

-0.13

POSITIVE LOGITS

how

0.31

 hearing

0.31

to

0.30

 knowing

0.27

 having

0.26

 being

0.25

 watching

0.25

 that

0.24

 seeing

0.23

 when

0.21

Activations Density 0.100%

emotional and positive expressions related to experiences

No Comments

No Known Activations