INDEX

Explanations

instances of repetition and routine in contexts related to learning or experiences

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

uars

-0.14

Ã¤hr

-0.14

inalg

-0.14

 Twice

-0.14

erspective

-0.14

é¡į

-0.14

imity

-0.13

à¹ģà¸ļ

-0.13

ONA

-0.13

Ø§ÙģØª

-0.13

POSITIVE LOGITS

 again

0.47

ad

0.44

again

0.43

 time

0.39

 Again

0.38

Again

0.37

 over

0.33

 ÑģÐ½Ð¾Ð²Ð°

0.32

åĨį

0.28

_again

0.28

Activations Density 0.156%

instances of repetition and routine in contexts related to learning or experiences

No Comments

No Known Activations