INDEX

Explanations

legal references and tax-related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

reuse

-0.15

avor

-0.15

 ç±

-0.14

zk

-0.14

inside

-0.14

iper

-0.14

_REUSE

-0.13

ayet

-0.13

 reels

-0.13

 personals

-0.13

POSITIVE LOGITS

 Income

0.30

 income

0.29

 asses

0.28

 Assess

0.26

Income

0.25

 taxpayer

0.23

CBD

0.23

asses

0.23

 taxpayers

0.23

 incomes

0.23

Activations Density 0.004%

legal references and tax-related terms

No Comments

No Known Activations