INDEX

Explanations

phrases highlighting contradictions or contrasting ideas

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

followed by time-related prepositions/adverbs

np_acts-logits-general · gemini-2.0-flash

recent or since

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Tomorrow

-0.66

 tomorrow

-0.63

tomorrow

-0.63

 domani

-0.62

 propOrder

-0.56

 Tomorrow

-0.54

 amanhã

-0.53

 nanti

-0.52

 Then

-0.52

onViewCreated

-0.50

POSITIVE LOGITS

 since

1.97

 lately

1.77

since

1.73

 SINCE

1.67

 recent

1.54

 depuis

1.44

Since

1.44

 recently

1.43

 fince

1.41

recent

1.41

Activations Density 0.664%

phrases highlighting contradictions or contrasting ideas

followed by time-related prepositions/adverbs

recent or since

No Comments

No Known Activations