INDEX

Explanations

numerical patterns and structured data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 QtGui

-0.57

makeText

-0.57

SpringBootTest

-0.55

énario

-0.50

区

-0.49

 zove

-0.49

enumii

-0.49

愧

-0.49

propTypes

-0.48

Predecesor

-0.48

POSITIVE LOGITS

GEBURTSDATUM

0.77

StoryboardSegue

0.68

WriteTagHelper

0.67

 ainfi

0.67

URLException

0.67

astéroïdes

0.66

 bahawa

0.60

 архивлан

0.59

 democrática

0.57

principalColumn

0.57

Activations Density 0.091%

numerical patterns and structured data

No Comments

No Known Activations