INDEX

Explanations

numerical data points related to energy production and capacity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 mill

-0.16

Vak

-0.15

 Pill

-0.15

ãĥ¬ãĥĥãĥĪ

-0.15

mL

-0.15

 grams

-0.15

 milli

-0.14

iband

-0.14

andy

-0.14

itty

-0.14

POSITIVE LOGITS

Meg

0.51

meg

0.48

gig

0.45

MW

0.43

Meg

0.43

 Mega

0.39

Gig

0.38

meg

0.38

MW

0.38

 mega

0.31

Activations Density 0.031%

numerical data points related to energy production and capacity

No Comments

No Known Activations