INDEX

Explanations

specific references to individual contexts or cases within the framework of human activity or studies

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

Pautan

-0.49

 censiti

-0.49

ьаж

-0.48

 Bare

-0.47

 estratégico

-0.47

 barefoot

-0.45

 transparencia

-0.44

 Naked

-0.44

Naked

-0.44

 estratég

-0.43

POSITIVE LOGITS

 individual

1.15

individual

1.07

Individual

1.06

 Individual

1.02

 individually

1.02

 INDIVIDUAL

0.98

 individuale

0.97

 individuales

0.91

 private

0.85

 individuel

0.85

Activations Density 0.954%

specific references to individual contexts or cases within the framework of human activity or studies

No Comments

No Known Activations