INDEX

Explanations

references to individuals in directorial or leadership positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

osci

-0.18

Ð»Ñİ

-0.17

ingly

-0.15

bish

-0.15

ICATION

-0.15

emit

-0.14

ÏģÎ¿Î¹

-0.14

ocker

-0.14

oj

-0.14

uka

-0.14

POSITIVE LOGITS

ship

0.30

ial

0.30

ate

0.27

ivity

0.27

ships

0.24

ially

0.23

ates

0.22

orate

0.22

ium

0.20

-general

0.19

Activations Density 0.054%

references to individuals in directorial or leadership positions

No Comments

No Known Activations