INDEX

Explanations

pronouns and expressions related to self-identity and interpersonal perspectives

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

 Ð¾Ð±ÑĢÐ°ÑĤ

-0.15

idor

-0.15

astle

-0.14

Ð¾Ð»ÑĮÐ·Ð¾Ð²

-0.14

 fallen

-0.14

LTR

-0.14

CLS

-0.14

anson

-0.14

igner

-0.13

Prostit

-0.13

POSITIVE LOGITS

 accomplished

0.32

 accomplish

0.29

 accompl

0.28

 doing

0.27

 done

0.23

accom

0.22

doing

0.22

 DONE

0.21

 Doing

0.21

 ÑģÐ¾Ð±Ð¾Ð¹

0.21

Activations Density 0.200%

pronouns and expressions related to self-identity and interpersonal perspectives

No Comments

No Known Activations