INDEX

Explanations

expressions related to observation or noticing actions

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

Follows "see," "saw," or "heard."

np_acts-logits-general · gemini-2.0-flash

seeing little or many

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

 ويكيميديا

-0.56

complexType

-0.53

 existencia

-0.53

 dicendo

-0.52

 להיות

-0.52

 existence

-0.51

 Tôi

-0.51

existence

-0.49

ấn

-0.49

findpost

-0.48

POSITIVE LOGITS

 referenties

0.70

icordia

0.68

how

0.66

RuleContext

0.63

rungsseite

0.63

 evidence

0.62

KommentareTeilen

0.61

 glimpses

0.59

 someone

0.58

 Walkover

0.58

Activations Density 0.146%

expressions related to observation or noticing actions

Follows "see," "saw," or "heard."

seeing little or many

No Comments

No Known Activations