INDEX

Explanations

legal terms and trademarks associated with products and companies

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

rott

-0.17

 Tart

-0.16

opian

-0.15

 lodging

-0.15

ãģĲ

-0.15

ÑĢÐ¾Ñģ

-0.15

 Crossing

-0.15

 ÑĢÐµÑģ

-0.14

 underground

-0.14

 bookstore

-0.14

POSITIVE LOGITS

 Dove

0.25

 Suns

0.25

 deter

0.24

 Magnum

0.20

 Lever

0.20

 Hind

0.19

ove

0.18

 Surf

0.18

 laundry

0.18

 detergent

0.17

Activations Density 0.030%

legal terms and trademarks associated with products and companies

No Comments

No Known Activations