INDEX

Explanations

mentions of Venice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

oire

-0.21

çľł

-0.15

ieved

-0.15

geries

-0.15

eru

-0.15

TOTYPE

-0.14

iras

-0.14

teen

-0.14

aÄŁa

-0.14

tsky

-0.14

POSITIVE LOGITS

Ven

0.23

ven

0.22

Ven

0.21

ereal

0.20

uez

0.18

ues

0.17

kat

0.17

ven

0.15

efore

0.15

mos

0.15

Activations Density 0.010%

mentions of Venice

No Comments

No Known Activations