INDEX

Explanations

academic affiliations and educational backgrounds

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

Ø§ÛĮÙĨ

-0.16

iag

-0.15

Contours

-0.15

 Basket

-0.15

pha

-0.15

olik

-0.15

ĮĢ

-0.14

Alv

-0.14

_queries

-0.13

akah

-0.13

POSITIVE LOGITS

 Cambridge

0.21

 Pennsylvania

0.21

 California

0.20

 Princeton

0.20

 Michigan

0.19

 Chicago

0.19

Tue

0.19

 Oxford

0.18

 Toronto

0.18

 Illinois

0.17

Activations Density 0.048%

academic affiliations and educational backgrounds

No Comments

No Known Activations