INDEX

Explanations

words and phrases related to subscriptions and suburban contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

obl

-0.18

ython

-0.17

izr

-0.16

stride

-0.16

Ð½Ð¾ÑģÑĤ

-0.16

chers

-0.15

ighton

-0.15

fully

-0.15

obre

-0.15

witter

-0.14

POSITIVE LOGITS

=sub

0.26

(Sub

0.24

/Sub

0.24

/sub

0.23

stract

0.22

tember

0.20

-Saharan

0.20

utex

0.19

mers

0.19

ively

0.19

Activations Density 0.074%

words and phrases related to subscriptions and suburban contexts

No Comments

No Known Activations