INDEX

Explanations

words and phrases related to action and accountability in a collaborative context

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

states or outcomes

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

 Мексичка

-0.37

 argint

-0.36

뀐

-0.35

'},

-0.35



-0.34

дыду

-0.34

},

-0.34

 doccia

-0.33

 препратки

-0.33

POSITIVE LOGITS

IntoConstraints

0.73

SaveChangesAsync

0.59

 يتيمه

0.57

herself

0.57

存于互联网档案馆

0.57

nonatomic

0.56

 arşivlendi

0.55

 yourself

0.54

ไว้

0.54

yourself

0.51

Activations Density 0.046%

words and phrases related to action and accountability in a collaborative context

states or outcomes

No Comments

No Known Activations