INDEX

Explanations

punctuation marks

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

ArrowToggle

-0.93

 propOrder

-0.92

oneofs

-0.84

ClientSize

-0.83

msgTypes

-0.79

 kaarangay

-0.78

aarrggbb

-0.78

elemField

-0.76

 InputDecoration

-0.73

 Biôgrafia

-0.73

POSITIVE LOGITS

 establecidos

0.34

 Keuangan

0.32

 Class

0.32

Jäh

0.32

Class

0.30

 profissional

0.30

 propuesto

0.30

 darb

0.28

 bougies

0.28

 suun

0.28

Activations Density 0.000%

punctuation marks

No Comments

No Known Activations