INDEX

Explanations

phrases related to legal and eligibility requirements

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

\Html

-0.15

Î¸Î¹

-0.15

ÙĪØ¯

-0.14

 lesson

-0.14

 ìŀĦ

-0.14

uf

-0.14

alat

-0.14

ondon

-0.14

 Biol

-0.14

hardt

-0.13

POSITIVE LOGITS

 eligibility

0.26

 eligible

0.25

 Elig

0.23

eligible

0.23

 qualify

0.22

 qualifying

0.21

elig

0.20

ellig

0.20

 qualifies

0.20

 entitled

0.19

Activations Density 0.111%

phrases related to legal and eligibility requirements

No Known Activations