INDEX

Explanations

references to specific locations within Boston

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 Cats

-0.21

NJ

-0.19

 Connecticut

-0.18

-0.17

anken

-0.17

 Peek

-0.17

NJ

-0.17

 Delaware

-0.16

.gwt

-0.16

/Main

-0.16

POSITIVE LOGITS

 Boston

0.36

Boston

0.32

oston

0.27

0.26

 Cambridge

0.26

 Harvard

0.23

MIT

0.23

 Celtics

0.22

 Massachusetts

0.21

fen

0.21

Activations Density 0.163%

references to specific locations within Boston

No Comments

No Known Activations