INDEX

Explanations

phrases that emphasize negation or contrast

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 Alembic

-0.75

 Vikipedi

-0.70

 firebaseConfig

-0.70

AsUp

-0.69

SBATCH

-0.66

 ProductService

-0.65

dotenv

-0.65

 BitmapFactory

-0.63

 comets

-0.63

Parcelize

-0.63

POSITIVE LOGITS

不仅仅

0.58

 merely

0.55

awaiter

0.54

 solely

0.51

 multiple

0.50

 mere

0.50

光是

0.48

不止

0.47

pre

0.46

Зноскі

0.45

Activations Density 0.166%

phrases that emphasize negation or contrast

No Comments

No Known Activations