INDEX

Explanations

references to historical artifacts and items of significance

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

agara

-0.18

noinspection

-0.17

-0.15

 Outs

-0.15

-0.14

 exhaust

-0.14

ogo

-0.14

 consuming

-0.14

Ð°Ð½ÑģÐ¸

-0.14

POSITIVE LOGITS

 owned

0.27

 handed

0.26

 loan

0.26

 belonged

0.23

 Loan

0.22

Owned

0.22

 confiscated

0.21

owned

0.21

 kept

0.21

loan

0.21

Activations Density 0.137%

references to historical artifacts and items of significance

No Comments

No Known Activations