INDEX

Explanations

references to educational resources and tools

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

âĢ¦↵

-0.27

âĢ¦”

-0.24

âĢ¦and

-0.23

âĢ¦

-0.22

âĢ¦"

-0.21

 âĢ¦↵

-0.21

 [âĢ¦]↵

-0.21

âĢ¦.

-0.20

âĢ¦the

-0.19

âĢ¦I

-0.19

POSITIVE LOGITS

#ab

0.16

#ad

0.16

#af

0.16

#ac

0.15

#aa

0.14

/***/

0.14

)application

0.14

/******/

0.13

)did

0.13

)frame

0.12

Activations Density 94.350%

references to educational resources and tools

No Comments

No Known Activations