INDEX

Explanations

references to specific individuals or unique identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

áº¡ch

-0.15

ncy

-0.14

åħ¬åħ±

-0.13

 trÃŃ

-0.13

Duc

-0.13

amarin

-0.13

mÃŃ

-0.13

 GPLv

-0.13

Ð°Ð¼Ð¿

-0.13

 duke

-0.12

POSITIVE LOGITS

 Moon

0.36

 Alpha

0.31

Moon

0.29

 moon

0.26

Alpha

0.25

 Voyager

0.25

 ALPHA

0.25

 astronauts

0.24

 lunar

0.24

 Apollo

0.23

Activations Density 0.005%

references to specific individuals or unique identifiers

No Comments

No Known Activations