INDEX

Explanations

numerical values indicating percentages or scores

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

underwater

-0.50

consul

-0.50

ConverterFactory

-0.49

Buona

-0.48

Consul

-0.47

 समीक्षाएं

-0.46

DataPropertyName

-0.46

Buddha

-0.45

 fune

-0.45

 Cowell

-0.45

POSITIVE LOGITS

mpe

0.54

titleMargin

0.53

TINGS

0.53

 Sixteenth

0.50

 Terrence

0.49

 Marburg

0.48

teen

0.47

istani

0.47

paris

0.47

EEN

0.47

Activations Density 0.889%

numerical values indicating percentages or scores

No Comments

No Known Activations