INDEX

Explanations

office-related terms and phrases indicating new developments or transitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

osoba

-0.42

 vainilla

-0.38

WriteTagHelper

-0.37

 Vogt

-0.36

 zvíř

-0.35

 labios

-0.35

InjectAttribute

-0.35

 pegatina

-0.34

val

-0.34

 manipulate

-0.33

POSITIVE LOGITS

 headquarters

0.69

 offices

0.67

 premises

0.67

 leased

0.65

 rented

0.65

 occupancy

0.63

offices

0.63

 Houſe

0.62

 occupied

0.62

 occupies

0.61

Activations Density 0.271%

office-related terms and phrases indicating new developments or transitions

No Comments

No Known Activations

office-related terms and phrases indicating new developments or transitions

No Comments

No Known Activations