INDEX

Explanations

temporal expressions related to duration and follow-up periods in treatments or studies

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

lengths of time (weeks, months)

np_acts-logits-general · gemini-2.0-flash

weeks and months

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 expresión

-0.38

stagland

-0.38

nearest

-0.35

 esquina

-0.35

 representado

-0.35

 niebla

-0.34

 árvore

-0.34

OfWeek

-0.34

 oreja

-0.33

 película

-0.33

POSITIVE LOGITS

 months

1.07

Months

1.00

 Months

0.96

months

0.96

 MONTHS

0.88

半年

0.84

 maanden

0.83

 Monate

0.82

 myſelf

0.78

 meses

0.77

Activations Density 0.256%

temporal expressions related to duration and follow-up periods in treatments or studies

lengths of time (weeks, months)

weeks and months

No Comments

No Known Activations