INDEX

Explanations

references to filling out forms or applications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

kas

-0.16

ä¹ĭ

-0.16

amo

-0.16

esting

-0.15

osto

-0.15

egas

-0.15

ystack

-0.14

 prime

-0.14

ihat

-0.14

pec

-0.14

POSITIVE LOGITS

 form

0.28

 fields

0.28

 Fields

0.24

Fields

0.22

fields

0.21

.fields

0.21

-fields

0.21

form

0.21

_FIELDS

0.21

 online

0.20

Activations Density 0.019%

references to filling out forms or applications

No Comments

No Known Activations