INDEX

Explanations

phrases that denote alternative names or titles for entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

hest

-0.16

irit

-0.15

ÄĲT

-0.15

 ØªØ±ÛĮÙĨ

-0.14

ICON

-0.14

NESS

-0.14

ãĥ¼ãĥĬ

-0.14

ames

-0.14

ë³µì§Ģ

-0.14

pyx

-0.14

POSITIVE LOGITS

 simply

0.27

as

0.22

 sebagai

0.20

 Simply

0.19

Simply

0.18

 either

0.17

 simplement

0.17

 jako

0.17

ä½ľ

0.16

als

0.16

Activations Density 0.030%

phrases that denote alternative names or titles for entities

No Comments

No Known Activations