INDEX

Explanations

phrases indicating statistical significance or results in a data-driven context

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

ьаж

-0.82

 otomatig

-0.80

 autorytatywna

-0.79

OGND

-0.77

хьтан

-0.73

ScopeManager

-0.71

resave

-0.71

quelize

-0.69

 Normdatei

-0.68

 EconPapers

-0.67

POSITIVE LOGITS

 shown

0.40

曖昧さ回避

0.32

 https

0.30

shown

0.29

 seen

0.29

image

0.29

面

0.28

 Arki

0.28

 montré

0.28

 picture

0.28

Activations Density 0.086%

phrases indicating statistical significance or results in a data-driven context

No Comments

No Known Activations