INDEX

Explanations

attributes and qualities that describe individuals, especially in the context of their personalities and performances

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

-0.16

ÏģÎºÎµ

-0.15

Ð·Ð½Ð°Ñĩ

-0.15

-0.14

.shapes

-0.14

ÑĸÑģÐ½Ð¾

-0.14

asty

-0.13

gos

-0.13

POSITIVE LOGITS

 personality

0.28

 manner

0.28

 presence

0.27

 infectious

0.25

 sense

0.23

presence

0.22

 Presence

0.22

 demeanor

0.21

 contagious

0.21

 voice

0.20

Activations Density 0.217%

attributes and qualities that describe individuals, especially in the context of their personalities and performances

No Comments

No Known Activations