INDEX

Explanations

dates and temporal references in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_3/width_16k/average_l0_90

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

LLocation

-0.57

 للاسماء

-0.53

ſelves

-0.52

WebVitals

-0.51

 ſta

-0.50

 Anſ

-0.49

continent

-0.49

ſelf

-0.48

Efq

-0.48

 indian

-0.47

POSITIVE LOGITS

 March

0.98

 march

0.94

 MARCH

0.94

March

0.94

MARCH

0.81

march

0.74

 marching

0.71

 marches

0.68

 marcha

0.67

 marzo

0.65

Activations Density 0.056%

dates and temporal references in the text

No Comments

No Known Activations