INDEX

Explanations

references to locations and incidents related to accidents or emergencies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

POCH

-0.18

 lÃ¶

-0.16

çĲ

-0.15

Äħ

-0.15

ccak

-0.15

Initializer

-0.14

vero

-0.14

@student

-0.14

 rect

-0.14

irsch

-0.14

POSITIVE LOGITS

 Oregon

0.35

Oregon

0.31

 Rogue

0.28

 Portland

0.27

 Eugene

0.26

Portland

0.26

0.25

 Bend

0.21

regon

0.20

 Salem

0.20

Activations Density 0.022%

references to locations and incidents related to accidents or emergencies

No Comments

No Known Activations