INDEX

Explanations

phrases expressing strong opinions or emotions about 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

 both

-0.24

 some

-0.24

 what

-0.20

 several

-0.20

 most

-0.20

 whatever

-0.20

both

-0.19

 somewhat

-0.18

 such

-0.18

 whichever

-0.17

POSITIVE LOGITS

like

0.33

itra

0.27

0.26

an

0.25

 thing

0.23

 things

0.20

-and

0.20

eta

0.19

-like

0.19

thing

0.18

Activations Density 0.072%

phrases expressing strong opinions or emotions about experiences

No Comments

No Known Activations