INDEX

Explanations

references to specific figures and data in scientific studies

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

Appears before labels (a, b, c, etc.) in figures

np_acts-logits-general · gemini-2.0-flash

figure labels (a, A)

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

static

-0.89

 static

-0.87

 الحره

-0.81

 dynamique

-0.72

dynamic

-0.67

 dynamic

-0.67

Geplaatst

-0.65

 dinámica

-0.65

 ویکی‌پدی

-0.64

 dinámico

-0.60

POSITIVE LOGITS

 RequestMethod

0.48

ofür

0.47

MLLoader

0.47

 tqdm

0.45

loài

0.44

EndInit

0.43

 QtWidgets

0.43

 GenerationType

0.41

ViewFeatures

0.41

omitempty

0.41

Activations Density 0.575%

references to specific figures and data in scientific studies

Appears before labels (a, b, c, etc.) in figures

figure labels (a, A)

No Comments

No Known Activations