INDEX

Explanations

phrases discussing opposing viewpoints or dual concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ucch

-0.15

ErrorException

-0.15

avit

-0.14

 æĻ´

-0.14

_sid

-0.14

-prepend

-0.14

å¡ļ

-0.14

loÅ¾

-0.14

ibus

-0.13

indsay

-0.13

POSITIVE LOGITS

 spectrum

0.44

 coin

0.44

coin

0.36

 Coin

0.33

 equation

0.32

 aisle

0.32

 Spectrum

0.31

 coins

0.30

Coin

0.29

 spectra

0.28

Activations Density 0.033%

phrases discussing opposing viewpoints or dual concepts

No Comments

No Known Activations