INDEX

Explanations

numerical measurements related to area and size

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

uos

-0.15

 destinationViewController

-0.15

illions

-0.15

emploi

-0.14

une

-0.14

 billions

-0.14

iert

-0.14

 Neighborhood

-0.14

atz

-0.14

oci

-0.14

POSITIVE LOGITS

sq

0.25

 square

0.25

-square

0.20

square

0.19

ft

0.18

sq

0.18

 feet

0.18

ft

0.18

 Square

0.17

_sq

0.17

Activations Density 0.031%

numerical measurements related to area and size

No Comments

No Known Activations