INDEX

Explanations

references to specific TV series and their key details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ken

-0.17

ÄĽn

-0.16

orda

-0.15

ixe

-0.15

å¦ĸ

-0.15

Acts

-0.15

anca

-0.15

endir

-0.14

thora

-0.14

icerca

-0.14

POSITIVE LOGITS

 meth

0.34

meth

0.28

 Breaking

0.28

 Meth

0.26

Breaking

0.25

 Albuquerque

0.25

 METH

0.21

 Walt

0.21

 Walter

0.20

 launder

0.19

Activations Density 0.004%

references to specific TV series and their key details

No Comments

No Known Activations