INDEX

Explanations

punctuation marks and numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

 Ð³Ð»Ñĥ

-0.15

UnderTest

-0.15

_Impl

-0.14

ighton

-0.14

amation

-0.14

hawk

-0.13

count

-0.13

å¼ı

-0.13

à¥įà¤¹

-0.13

 truyá»ģn

-0.13

POSITIVE LOGITS

 Woman

0.19

Woman

0.15

osi

0.15

esis

0.14

ox

0.14

 Labs

0.14

ugo

0.14

iod

0.14

 poster

0.14

olla

0.14

Activations Density 0.008%

punctuation marks and numerical values

No Comments

No Known Activations