INDEX

Explanations

instances of the word "the" as well as variations of emphasis and context therein

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

anford

-0.17

tatus

-0.15

ober

-0.15

achel

-0.15

ught

-0.14

 McKenzie

-0.14

combe

-0.14

Ãºc

-0.14

ypse

-0.14

paginate

-0.13

POSITIVE LOGITS

 strm

0.16

 divert

0.15

 RedirectTo

0.14

ladu

0.14

ajo

0.13

eny

0.13

Ð¼ÐµÐ»ÑĮ

0.13

erken

0.13

agem

0.13

Activations Density 0.292%

instances of the word "the" as well as variations of emphasis and context therein

No Comments

No Known Activations