INDEX

Explanations

sections and labels in a document

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

asma

-0.16

ole

-0.15

avig

-0.14

aÃ§

-0.14

iges

-0.14

ibri

-0.14

url

-0.14

 Blanch

-0.13

dej

-0.13

isma

-0.13

POSITIVE LOGITS

sec

0.26

 section

0.24

section

0.23

sec

0.22

SEC

0.22

Sec

0.22

.sec

0.21

Sec

0.21

.section

0.20

_section

0.19

Activations Density 0.048%

sections and labels in a document

No Comments

No Known Activations