INDEX

Explanations

geographical locations and their descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

 capitals

-0.17

holm

-0.16

 cities

-0.15

 Bever

-0.15

 Manila

-0.14

UBLE

-0.14

 Cities

-0.14

angkan

-0.14

amburger

-0.13

andum

-0.13

POSITIVE LOGITS

 Oregon

0.18

 Illinois

0.17

 Maryland

0.16

 Pennsylvania

0.16

 Ohio

0.16

 Kentucky

0.15

LETE

0.15

 Michigan

0.15

 Missouri

0.15

SE

0.15

Activations Density 0.133%

geographical locations and their descriptions

No Comments

No Known Activations