INDEX

Explanations

references to Harry Potter and its associated characters or elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

è¦

-0.17

osg

-0.16

_FMT

-0.16

ÙĪØ§Ùĩ

-0.16

èĴ

-0.16

CTL

-0.14

æłı

-0.14

Ã¢y

-0.14

omial

-0.14

POSITIVE LOGITS

 Sirius

0.25

Ron

0.24

 Moody

0.24

Cho

0.22

Ced

0.22

 Dudley

0.21

 Ginny

0.21

Cho

0.20

McG

0.20

 Chamber

0.20

Activations Density 0.005%

references to Harry Potter and its associated characters or elements

No Comments

No Known Activations