INDEX

Explanations

terminology related to debate and competitive debate formats

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

æ³½

-0.15

utors

-0.14

ÙĬÙĨÙĩ

-0.14

aker

-0.14

ucc

-0.13

ppard

-0.13

abase

-0.13

ÐµÐ¹Ð½

-0.13

ibble

-0.13

 arac

-0.13

POSITIVE LOGITS

 Open

0.32

 open

0.28

 Senior

0.28

 division

0.27

 divisions

0.27

age

0.26

Open

0.25

 senior

0.25

 OPEN

0.25

 ages

0.24

Activations Density 0.071%

terminology related to debate and competitive debate formats

No Comments

No Known Activations