INDEX

Explanations

references to personal relationships and experiences

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

shima

-0.15

wik

-0.15

pio

-0.15

idents

-0.14

matic

-0.14

cken

-0.14

eyJ

-0.14

ovsky

-0.13

utomation

-0.13

.staff

-0.13

POSITIVE LOGITS

 hosts

0.28

 surroundings

0.21

 visit

0.21

 host

0.20

hosts

0.20

compan

0.19

visit

0.19

 companions

0.19

 stay

0.17

 destination

0.17

Activations Density 0.185%

references to personal relationships and experiences

No Comments

No Known Activations