INDEX

Explanations

mathematical terms and concepts related to geometry and spatial dimensions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 circular

-0.20

 circle

-0.19

 ellipse

-0.19

 Circle

-0.19

circle

-0.18

 hors

-0.18

 Circular

-0.16

 circles

-0.16

Circle

-0.16

Circular

-0.16

POSITIVE LOGITS

 Cube

0.43

 cube

0.43

cub

0.41

 cubes

0.40

Cube

0.38

cube

0.38

Cub

0.37

_cube

0.36

(cube

0.35

 ÐºÑĥÐ±

0.33

Activations Density 0.023%

mathematical terms and concepts related to geometry and spatial dimensions

No Comments

No Known Activations