INDEX

Explanations

references to Iranian culture and history

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

ッキー

-0.27

öße

-0.26

 caribe

-0.25

キャベツ

-0.25

Beep

-0.25

 taches

-0.24

 crossorigin

-0.24

nsf

-0.24

沖縄

-0.24

 aura

-0.24

POSITIVE LOGITS

 Iranian

1.06

 Tehran

1.05

 Iran

1.02

 Persian

0.97

Iran

0.91

Iranian

0.91

 iran

0.88

Persian

0.88

 Persia

0.87

伊朗

0.85

Activations Density 0.038%

references to Iranian culture and history

No Comments

No Known Activations