INDEX

Explanations

references to time and duration, particularly concerning the past and events that have occurred

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

posedge

-0.69

ksikon

-0.67

 Winaray

-0.66

 Himo

-0.65

 bezeichneter

-0.65

bootstrapcdn

-0.63

//

-0.62

Obrázky

-0.60

UnsafeEnabled

-0.60

имущества

-0.58

POSITIVE LOGITS

 earlier

0.55

昔

0.53

 vroeger

0.53

 previous

0.53

 vorige

0.52

 السابق

0.52

 прошло

0.50

 früheren

0.49

 ursprünglichen

0.49

 Autre

0.49

Activations Density 0.365%

references to time and duration, particularly concerning the past and events that have occurred

No Comments

No Known Activations