INDEX

Explanations

time-related expressions and scheduling information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

存于互联网档案馆

-0.94

 فريبيس

-0.89

 Савезне

-0.85

protoimpl

-0.78

 متعلقه

-0.74

ніципалі

-0.71

 ostavi

-0.71

 disambiguazione

-0.69

تقاوى

-0.67

Jereo

-0.66

POSITIVE LOGITS

 noon

1.59

1.28

 midday

1.23

 Noon

1.22

 midnight

1.08

 lunchtime

1.07

noon

1.04

Noon

1.00

 lunch

0.86

midnight

0.80

Activations Density 0.128%

time-related expressions and scheduling information

No Comments

No Known Activations