INDEX

Explanations

references to discounts and price reductions

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

discount

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

frauen

-0.47

<em>

-0.46

Here

-0.45

 feminina

-0.44

 femininas

-0.43

 Rowley

-0.43

)[

-0.43

here

-0.42

Wo

-0.42

-0.41

POSITIVE LOGITS

 Discount

1.24

Discount

1.10

 discount

1.10

discount

1.02

 Discounts

0.94

 discounts

0.92

 discounting

0.91

 descuento

0.77

Discounts

0.77

 discounted

0.76

Activations Density 0.002%

references to discounts and price reductions

discount

No Comments

No Known Activations

references to discounts and price reductions

discount

No Comments

No Known Activations