INDEX

Explanations

references to pop culture icons and their projects

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

TECTED

-0.17

ropolis

-0.16

rosse

-0.16

Ð¾Ð»Ð¾ÑĤ

-0.16

POSIT

-0.15

rof

-0.15

zas

-0.14

åħĥ

-0.14

 Labrador

-0.14

pas

-0.14

POSITIVE LOGITS

Bey

0.37

bey

0.31

Jay

0.31

Jay

0.28

 Lemon

0.28

 Formation

0.28

 Know

0.24

 Destiny

0.24

Formation

0.24

 Sasha

0.23

Activations Density 0.014%

references to pop culture icons and their projects

No Comments

No Known Activations