INDEX

Explanations

terms related to mathematical and computational frameworks

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

ASN

-0.15

oga

-0.15

 Ther

-0.15

meteor

-0.15

-0.14

 swear

-0.14

anton

-0.14

 Meteor

-0.14

Å¡t

-0.14

POSITIVE LOGITS

Cab

0.29

 mixing

0.29

Cab

0.28

CP

0.26

 Mixing

0.25

CK

0.25

mix

0.23

CP

0.23

Mix

0.23

CK

0.22

Activations Density 0.001%

terms related to mathematical and computational frameworks

No Comments

No Known Activations