INDEX

Explanations

references to the Internet and its applications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

Ko

-0.18

bsites

-0.15

ught

-0.15

ÑĤÐ¾ÑĢ

-0.15

etur

-0.14

ko

-0.14

ries

-0.14

umont

-0.14

itis

-0.13

ettes

-0.13

POSITIVE LOGITS

mÃ©na

0.16

ohana

0.15

 KrÃ¡l

0.15

bron

0.14

nodoc

0.14

tin

0.14

çĿ£

0.14

alat

0.14

 brass

0.14

Ø¹ÙĬØ©

0.14

Activations Density 0.012%

references to the Internet and its applications

No Comments

No Known Activations