INDEX

Explanations

references to archived digital content and specific dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

oyo

-0.16

stringValue

-0.15

abal

-0.15

rips

-0.14

Ð¾ÑĤÐ°

-0.14

emem

-0.14

asti

-0.14

vents

-0.14

OTO

-0.14

ipro

-0.13

POSITIVE LOGITS

oodle

0.18

sted

0.17

agli

0.16

 Annunci

0.16

rette

0.15

ï¼ĮåŃĺäºİ

0.15

Î¼Î¹Î¿

0.15

-LAST

0.15

 Bever

0.15

 Bedford

0.15

Activations Density 0.002%

references to archived digital content and specific dates

No Comments

No Known Activations

references to archived digital content and specific dates

No Comments

No Known Activations