INDEX

Explanations

punctuation marks, particularly those that denote emphasis or action

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

ikon

-0.15

#ac

-0.15

heimer

-0.15

ecure

-0.15

ìĨ

-0.14

mÄĽr

-0.14

 Affero

-0.14

 Prod

-0.14

 âĹĦ

-0.14

ccd

-0.14

POSITIVE LOGITS

Copyright

0.22

 Copyright

0.18

0.16

ousse

0.15

(*)

0.15

 Greenwood

0.14

 Simpson

0.14

(ii

0.14

 scrim

0.14

ooo

0.14

Activations Density 0.986%

punctuation marks, particularly those that denote emphasis or action

No Comments

No Known Activations