INDEX

Explanations

specific mentions of names or key figures

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

names of people and places

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

 informée

-0.82

endphp

-0.78

WithIOException

-0.68

OGND

-0.67

MessageOf

-0.63

andExpect

-0.61

makeText

-0.60

хьтан

-0.59

 Мексичка

-0.57

 Allociné

-0.56

POSITIVE LOGITS

Sooo

0.41

 Strategic

0.40

愷

0.40

Sol

0.39

lut

0.38

 Register

0.38

Dia

0.38

 Hard

0.38

straw

0.38

 Continuum

0.37

Activations Density 0.102%

specific mentions of names or key figures

names of people and places

No Comments

No Known Activations