INDEX

Explanations

references to programming concepts and data structure elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 const

-0.23

const

-0.16

ume

-0.15

	const

-0.15

amage

-0.14

)const

-0.14

::

-0.14

,const

-0.14

::_

-0.14

(const

-0.14

POSITIVE LOGITS

 final

0.29

(boolean

0.26

(final

0.26

final

0.25

boolean

0.25

 boolean

0.24

(String

0.20

 String

0.20

 finally

0.20

 throws

0.19

Activations Density 0.029%

references to programming concepts and data structure elements

No Comments

No Known Activations