INDEX

Explanations

references to rankings or achievements in a competitive context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

sons

-0.19

 sons

-0.17

wayne

-0.17

alf

-0.17

 Sons

-0.16

son

-0.16

SON

-0.16

 Mark

-0.16

Mark

-0.15

alim

-0.15

POSITIVE LOGITS

 Haley

0.24

Bri

0.23

 Morgan

0.23

 Cassidy

0.22

 Madison

0.21

 Paige

0.21

 Hannah

0.20

 Mack

0.19

 Jord

0.19

morgan

0.19

Activations Density 0.085%

references to rankings or achievements in a competitive context

No Comments

No Known Activations