INDEX

Explanations

instructions related to organization and cleanup tasks

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

 reopened

-0.15

 Rede

-0.14

iola

-0.14

oder

-0.14

unger

-0.14

itted

-0.14

ONY

-0.14

uÄį

-0.14

ÑĨÐµ

-0.14

 reopening

-0.13

POSITIVE LOGITS

 clean

0.31

 iron

0.28

 polish

0.27

tid

0.27

 tidy

0.26

 Clean

0.25

clean

0.24

 fresh

0.24

 fine

0.23

 smooth

0.23

Activations Density 0.187%

instructions related to organization and cleanup tasks

No Comments

No Known Activations

instructions related to organization and cleanup tasks

No Comments

No Known Activations