INDEX

Explanations

references to museums and their features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

Ú¯Ø±ÛĮ

-0.16

ombo

-0.15

è³

-0.15

 trá»£

-0.14

encode

-0.14

 Apost

-0.14

ë³¸

-0.14

 airports

-0.14

obuf

-0.14

ausible

-0.13

POSITIVE LOGITS

 displays

0.25

 exhibits

0.25

 Displays

0.21

/display

0.21

 collections

0.20

 exhibit

0.20

Displays

0.20

 displayed

0.20

 museum

0.20

 display

0.20

Activations Density 0.136%

references to museums and their features

No Comments

No Known Activations