INDEX

Explanations

features and specifications related to technology and gadgets

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 tarihli

-0.15

 Vapor

-0.14

 prive

-0.14

tÃŃ

-0.14

ibur

-0.14

-desktop

-0.14

 baise

-0.14

ulado

-0.13

itous

-0.13

IRON

-0.13

POSITIVE LOGITS

car

0.26

 Navigation

0.24

 Sirius

0.23

/navigation

0.22

Car

0.20

 navigation

0.19

nav

0.19

è½¦

0.18

Sir

0.18

Nav

0.18

Activations Density 0.026%

features and specifications related to technology and gadgets

No Comments

No Known Activations