INDEX

Explanations

patterns and discussions around attention and visibility in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

curring

-0.19

 ë°ľ

-0.15

ContentSize

-0.14

uario

-0.14

AMESPACE

-0.14

acin

-0.14

Bak

-0.13

ansi

-0.13

Ves

-0.13

 march

-0.13

POSITIVE LOGITS

OOD

0.14

erto

0.14

ToObject

0.14

ryo

0.14

 Theodore

0.14

logan

0.13

 æħ

0.13

ê¸Ī

0.13

Ø±Ùģ

0.13

 GOODS

0.13

Activations Density 0.155%

patterns and discussions around attention and visibility in various contexts

No Comments

No Known Activations