INDEX

Explanations

phrases that include the word "under" or related terms indicating position

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ãĥªãĤ«

-0.17

uno

-0.17

exus

-0.15

mall

-0.15

ectar

-0.15

Ã¼mÃ¼ÅŁ

-0.14

bach

-0.14

å±Ģ

-0.14

ss

-0.14

ERA

-0.14

POSITIVE LOGITS

neath

0.34

 covers

0.25

 hood

0.23

 Covers

0.22

 layers

0.22

 beneath

0.22

covers

0.20

hood

0.20

 cover

0.20

foot

0.20

Activations Density 0.035%

phrases that include the word "under" or related terms indicating position

No Comments

No Known Activations