INDEX

Explanations

phrases related to various jobs and work experiences

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

 Planner

-0.15

 negot

-0.14

åĬŁ

-0.14

å¤«äºº

-0.14

ncmp

-0.14

 Shopping

-0.13

raquo

-0.13

zk

-0.13

URRENCY

-0.13

´Ī

-0.13

POSITIVE LOGITS

 bart

0.29

 delivering

0.27

 wait

0.26

 sweeping

0.26

 deliveries

0.24

 delivery

0.24

odd

0.24

lif

0.24

 cler

0.22

 washing

0.22

Activations Density 0.194%

phrases related to various jobs and work experiences

No Comments

No Known Activations