INDEX

Explanations

phrases indicating relativity and comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

est

-0.25

 sharper

-0.24

 brightest

-0.22

 happiest

-0.21

 brighter

-0.21

 weakest

-0.21

 hottest

-0.20

 quickest

-0.19

EST

-0.19

 hardest

-0.19

POSITIVE LOGITS

wil

0.24

 tast

0.21

 loft

0.21

wil

0.20

 fatt

0.19

 pret

0.19

Wil

0.19

cra

0.19

 clums

0.18

ju

0.18

Activations Density 0.175%

phrases indicating relativity and comparisons

No Comments

No Known Activations