INDEX

Explanations

attributes related to influential figures in architecture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

новниш

-0.49

 Hopefully

-0.45

Autoritní

-0.44

吋

-0.44

ScopeManager

-0.44

AccessorTable

-0.41

 становника

-0.41

 noDo

-0.41

Hopefully

-0.39

突

-0.38

POSITIVE LOGITS

 decades

0.83

 pioneering

0.70

 yıllar

0.63

 décadas

0.60

decade

0.60

 longevity

0.59

 åren

0.56

 career

0.55

 lifetime

0.54

 lifelong

0.54

Activations Density 0.398%

attributes related to influential figures in architecture

No Comments

No Known Activations