INDEX

Explanations

words related to intelligence and integration

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

abouts

-0.17

IPH

-0.16

bsites

-0.15

ulace

-0.15

 serialVersionUID

-0.15

agas

-0.15

ilitation

-0.14

 ActionTypes

-0.14

istrovstvÃŃ

-0.14

geries

-0.14

POSITIVE LOGITS

ees

0.21

ted

0.21

lectual

0.18

apolis

0.18

abelle

0.17

uitive

0.17

ently

0.16

wine

0.15

å¤§åĪ©

0.15

uous

0.15

Activations Density 0.088%

words related to intelligence and integration

No Comments

No Known Activations