INDEX

Explanations

references to time intervals and durations

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_7/width_16k/average_l0_92

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 menik

-0.47

Зноскі

-0.46

astéroïdes

-0.46

 שוליים

-0.45

SourceChecksum

-0.45

 تضيفلها

-0.43

LabelTagHelper

-0.43

velope

-0.43

 aukš

-0.43

webElement

-0.42

POSITIVE LOGITS

 week

1.04

 days

0.95

三天

0.92

两天

0.90

 weeks

0.86

Days

0.85

 Days

0.82

几天

0.82

Week

0.81

幾天

0.81

Activations Density 0.252%

references to time intervals and durations

No Known Activations