INDEX

Explanations

references to the word "Pi" in various contexts

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

edir

-0.16

ilk

-0.15

amer

-0.15

ingham

-0.15

alars

-0.14

izoph

-0.14

ymm

-0.14

onymous

-0.14

clipse

-0.14

 stre

-0.14

POSITIVE LOGITS

LOT

0.23

aget

0.23

otr

0.20

ÂŁ

0.20

hole

0.19

aggio

0.19

wik

0.18

errick

0.17

-hole

0.17

holes

0.17

Activations Density 0.011%

references to the word "Pi" in various contexts

No Comments

No Known Activations