INDEX

Explanations

phrases indicating quantities or proportions in statistical contexts

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

"In/of" followed by a number

np_acts-logits-general · gemini-2.0-flash

every numerical proportion

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

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 насељу

-0.80

 للمعارف

-0.78

脚注の使い方

-0.77

ruptedException

-0.73

 ExecuteAsync

-0.73

 المعيارى

-0.73

Obrázky

-0.69

 "}";

-0.68

ativement

-0.66

 بيها

-0.64

POSITIVE LOGITS

 every

2.04

every

1.77

Every

1.68

 Every

1.67

 EVERY

1.19

 ogni

1.18

EVERY

1.17

 setiap

1.04

 каж

0.90

每

0.89

Activations Density 0.216%

phrases indicating quantities or proportions in statistical contexts

"In/of" followed by a number

every numerical proportion

No Comments

No Known Activations