INDEX

Explanations

programming-related terms and concepts, particularly those associated with function definitions and variable management

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

obble

-0.15

Tam

-0.15

egra

-0.15

WidthSpace

-0.15

iol

-0.15

 surface

-0.15

fov

-0.14

akat

-0.14

icina

-0.14

arkin

-0.14

POSITIVE LOGITS

wx

0.27

(wx

0.23

wx

0.22

Sizer

0.20

=wx

0.20

.wx

0.19

EVT

0.19

IDC

0.19

 wxDefault

0.19

(IDC

0.18

Activations Density 0.006%

programming-related terms and concepts, particularly those associated with function definitions and variable management

No Comments

No Known Activations