INDEX

Explanations

actions related to movement and interaction among characters

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

ynet

-0.19

wards

-0.17

 outward

-0.16

loo

-0.15

éĻµ

-0.14

Ø²Ø±

-0.14

opolitan

-0.14

eday

-0.14

astle

-0.13

oga

-0.13

POSITIVE LOGITS

 home

0.25

 back

0.25

 downstairs

0.24

 down

0.23

 inside

0.22

 outside

0.22

 downtown

0.22

 upstairs

0.22

 across

0.20

up

0.19

Activations Density 0.201%

actions related to movement and interaction among characters

No Comments

No Known Activations