INDEX

Explanations

phrases related to trust and transparency in voting systems

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 stag

-0.17

etine

-0.15

stag

-0.15

obil

-0.15

æ±ł

-0.14

illac

-0.14

 Pipes

-0.14

usercontent

-0.14

ett

-0.14

igon

-0.14

POSITIVE LOGITS

 touch

0.31

 touchscreen

0.30

 machines

0.29

touch

0.28

 paper

0.27

Die

0.27

tab

0.26

 Touch

0.26

Touch

0.25

-tab

0.25

Activations Density 0.045%

phrases related to trust and transparency in voting systems

No Comments

No Known Activations