INDEX

Explanations

references to bathrooms and showers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 مرئيه

-0.65

 Marcello

-0.56

BagLayout

-0.53

ьаж

-0.50

 للاسماء

-0.49

__::

-0.49

 grasshopper

-0.49

TargetException

-0.48

 dishonesty

-0.48

 RSSSF

-0.47

POSITIVE LOGITS

 shower

0.71

 bath

0.69

Shower

0.68

 Shower

0.67

浴

0.63

 bathtub

0.61

 bathroom

0.61

 bathing

0.61

shower

0.60

 bathe

0.60

Activations Density 0.070%

references to bathrooms and showers

No Comments

No Known Activations