INDEX

Explanations

names and titles of historical figures connected to nobility

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

yro

-0.16

 McKay

-0.16

 poil

-0.15

ILON

-0.14

Mc

-0.14

_mc

-0.14

utsch

-0.14

urai

-0.14

 Mickey

-0.14

 Kaiser

-0.14

POSITIVE LOGITS

 Vere

0.23

Vy

0.21

Pag

0.20

Pon

0.20

 Fans

0.20

 Vill

0.20

Nug

0.19

 Dorm

0.19

 Mont

0.19

 Hunger

0.19

Activations Density 0.055%

names and titles of historical figures connected to nobility

No Comments

No Known Activations