INDEX

Explanations

words that indicate suitability and appropriateness for specific purposes or contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

InvalidProtocol

-0.88

GraphicsUnit

-0.78

 ویکی‌پدی

-0.78

OGND

-0.76

GEBURTSDATUM

-0.71

 Италијани

-0.70

 يتيمه

-0.68

 Roskov

-0.67

BeginContext

-0.67

 насеље

-0.66

POSITIVE LOGITS

 standalone

0.90

 paired

0.89

 pairing

0.77

 alone

0.74

alone

0.72

 complement

0.70

 accompany

0.70

 Used

0.70

 used

0.69

Used

0.69

Activations Density 0.316%

words that indicate suitability and appropriateness for specific purposes or contexts

No Comments

No Known Activations