INDEX

Explanations

phrases indicating exceptions or limitations in eligibility or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

esco

-0.16

undi

-0.16

emes

-0.15

hong

-0.15

uggage

-0.14

å°Ķ

-0.14

gest

-0.14

 tÆ°Æ¡ng

-0.14

od

-0.14

zeug

-0.14

POSITIVE LOGITS

 exclusive

0.32

 limited

0.32

exclusive

0.27

limited

0.24

Exclusive

0.24

 exclusively

0.24

 exhausted

0.24

 Exclusive

0.23

 exhaustive

0.22

 restricted

0.22

Activations Density 0.007%

phrases indicating exceptions or limitations in eligibility or conditions

No Comments

No Known Activations