INDEX

Explanations

sections and subsections in a structured document

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

eniable

-0.16

reau

-0.15

abbo

-0.15

leigh

-0.14

iveau

-0.14

locker

-0.14

declspec

-0.14

herits

-0.14

fung

-0.14

caler

-0.14

POSITIVE LOGITS

gu

0.16

 Provid

0.15

 Tome

0.15

unda

0.14

onto

0.14

 Budd

0.14

 material

0.14

bek

0.14

Bek

0.14

oS

0.13

Activations Density 0.064%

sections and subsections in a structured document

No Comments

No Known Activations