INDEX

Explanations

references to New York or its neighborhoods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 Stick

-0.16

ainer

-0.15

editary

-0.15

 drum

-0.15

 blindly

-0.15

èģ¯

-0.15

ë¦Ħ

-0.14

 Coff

-0.14

 Manitoba

-0.14

.bz

-0.14

POSITIVE LOGITS

 York

0.32

 Haven

0.21

York

0.21

ìļķ

0.20

 YORK

0.20

 york

0.20

 Jersey

0.19

-Y

0.19

NY

0.18

 Roch

0.18

Activations Density 0.023%

references to New York or its neighborhoods

No Comments

No Known Activations