INDEX

Explanations

terms related to packing density and particle interactions in a system

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

 sign

-0.33

Expose

-0.33

 Mulder

-0.33

gdx

-0.31

leistung

-0.31

 gevol

-0.30

 steeds

-0.30

 acı

-0.30

 captured

-0.29

 built

-0.29

POSITIVE LOGITS

 BoxFit

0.63

OGND

0.63

 autorytatywna

0.58

principalColumn

0.57

 EconPapers

0.57

 Density

0.55

 betweenstory

0.55

Попис

0.54

 &___

0.54

تقاوى

0.54

Activations Density 0.109%

terms related to packing density and particle interactions in a system

No Comments

No Known Activations