INDEX

Explanations

references to significant buildings and architectural elements

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

хео

-0.38

Côte

-0.37

ătă

-0.37

 φυσ

-0.36

 Pascual

-0.35

webElementXpaths

-0.35

 Fuentes

-0.34

ankt

-0.33

ofern

-0.33

CreateIndex

-0.33

POSITIVE LOGITS

CloseOperation

0.90

 buildings

0.85

 building

0.82

 Buildings

0.76

building

0.76

expandindo

0.74

buildings

0.74

built

0.73

 edifice

0.73

 built

0.73

Activations Density 0.268%

references to significant buildings and architectural elements

No Comments

No Known Activations