INDEX

Explanations

terms related to preparation and organization

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

oa̍t

-0.65

onedDateTime

-0.58

 BorderSide

-0.57

 Nuss

-0.57

strokeWidth

-0.56

 clientId

-0.55

 menuItem

-0.55

calça

-0.55

 phoneNumber

-0.55

 OnClickListener

-0.54

POSITIVE LOGITS

 prepared

1.13

 prepare

1.13

prepare

1.06

 Prepare

1.05

 preparation

1.04

 prepar

0.99

prepared

0.99

Prepare

0.98

 Prepar

0.95

 Prepared

0.94

Activations Density 0.020%

terms related to preparation and organization

No Comments

No Known Activations