INDEX

Explanations

references to methods or processes related to improvement

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

Not in Any Lists

No Comments

Negative Logits

ç¥¨

-0.18

-scalable

-0.16

etat

-0.14

å«

-0.14

/wiki

-0.13

ì°¨

-0.13

plete

-0.13

 Bride

-0.13

 resourceName

-0.13

aginator

-0.13

POSITIVE LOGITS

 column

0.33

 Column

0.30

 columns

0.27

Column

0.26

column

0.26

 columnist

0.26

-column

0.25

 Opinion

0.25

 Columns

0.25

_column

0.24

Activations Density 0.204%

references to methods or processes related to improvement

No Comments

No Known Activations