INDEX

Explanations

terms related to interaction and engagement

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

venir

-0.18

Ñľ

-0.16

jours

-0.16

ego

-0.16

Ã±ana

-0.15

áºŃt

-0.15

ongs

-0.15

eno

-0.15

Ð½Ð³

-0.14

POSITIVE LOGITS

ives

0.21

ative

0.21

ivate

0.20

al

0.19

å¼ı

0.19

ype

0.18

uality

0.18

uator

0.18

iveness

0.18

AFX

0.18

Activations Density 0.015%

terms related to interaction and engagement

No Comments

No Known Activations