INDEX

Explanations

terms related to measurements or counts of entities

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

Quantity or count followed by "per"

np_acts-logits-general · gemini-2.0-flash

counts of categories

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

everywhere

-0.69

们

-0.64

 everywhere

-0.63

們

-0.61

 Everywhere

-0.61

们的

-0.59

 Theſe

-0.58

 Galen

-0.58

 Zeiten

-0.57

 utafitiHapana

-0.57

POSITIVE LOGITS

 apiece

0.91

 each

0.70

ImageContext

0.67

 plus

0.60

 total

0.59

per

0.58

 totalt

0.57

 PyTuple

0.56

each

0.56

initComponents

0.55

Activations Density 0.644%

terms related to measurements or counts of entities

Quantity or count followed by "per"

counts of categories

No Comments

No Known Activations