INDEX

Explanations

references to employment or job-related 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

xaf

-0.18

xdc

-0.16

egin

-0.15

.setViewport

-0.14

AndPassword

-0.14

åºĬ

-0.14

resident

-0.14

unken

-0.14

ours

-0.14

POSITIVE LOGITS

 full

0.25

odd

0.25

 part

0.25

 shifts

0.24

 graveyard

0.23

 side

0.19

 construction

0.19

 nights

0.18

 Grave

0.18

 strange

0.17

Activations Density 0.049%

references to employment or job-related activities

No Comments

No Known Activations