INDEX

Explanations

mathematical symbols and terms commonly used in equations or algorithms

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

ergus

-0.16

ania

-0.16

ignon

-0.16

uchen

-0.15

omed

-0.15

imus

-0.14

 omas

-0.14

 friction

-0.14

mug

-0.14

localized

-0.14

POSITIVE LOGITS

 codes

0.30

 Codes

0.26

 code

0.26

 Reed

0.24

codes

0.24

 Code

0.23

_codes

0.23

Code

0.23

cod

0.23

_code

0.22

Activations Density 0.009%

mathematical symbols and terms commonly used in equations or algorithms

No Comments

No Known Activations