INDEX

Explanations

references to San Francisco

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

logen

-0.16

dorf

-0.16

ypsum

-0.15

stdin

-0.15

ÑĥÐ»Ñİ

-0.15

lotte

-0.15

atsby

-0.15

ONO

-0.14

illard

-0.14

mobx

-0.14

POSITIVE LOGITS

 Francisco

0.40

Franc

0.31

 Mate

0.30

 franc

0.30

 Franc

0.29

 Fran

0.26

 Francis

0.25

 ÙģØ±Ø§ÙĨ

0.23

Mate

0.23

 mate

0.23

Activations Density 0.003%

references to San Francisco

No Comments

No Known Activations