INDEX

Explanations

terms related to representation or the act of representing

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

"representing" or "represented"

np_acts-logits-general · gemini-2.0-flash

is represented by

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 санк

-0.62

XtraEditors

-0.60

ikation

-0.59

omány

-0.58

воро

-0.57

 Civil

-0.55

 OkHttpClient

-0.54

Excessive

-0.54

chaikovsky

-0.53

 Nobel

-0.53

POSITIVE LOGITS

 representations

1.27

 Represents

1.27

 Representations

1.25

 representing

1.23

 Represent

1.22

Represents

1.22

Represent

1.21

 representation

1.19

 represents

1.18

 represen

1.16

Activations Density 0.138%

terms related to representation or the act of representing

"representing" or "represented"

is represented by

No Comments

No Known Activations