INDEX

Explanations

terms related to training and evaluation processes in a professional context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 Halk

-0.16

laÄį

-0.15

iegel

-0.15

 Malk

-0.15

scribed

-0.15

borg

-0.15

ideon

-0.15

çĥ

-0.15

antine

-0.15

Ã¡le

-0.14

POSITIVE LOGITS

 registrar

0.27

 registr

0.25

Registrar

0.23

 Consultant

0.23

 Registrar

0.22

 consultant

0.22

GMC

0.21

ana

0.21

 Specialty

0.20

FR

0.20

Activations Density 0.074%

terms related to training and evaluation processes in a professional context

No Comments

No Known Activations