INDEX

Explanations

the word "one"

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 disambiguazione

-0.95

曖昧さ回避

-0.87

 <<<<<<<<<<<<<<

-0.86

 Мексичка

-0.83

exitRule

-0.83

SharedCtor

-0.80

 betweenstory

-0.78

subpackage

-0.75

EDEFAULT

-0.72

yntaxException

-0.71

POSITIVE LOGITS

 einen

0.77

 einem

0.67

one

0.67

 miglior

0.60

 mejor

0.60

one

0.60

 Einen

0.58

 MEJOR

0.57

の一

0.54

 caccia

0.54

Activations Density 0.044%

the word "one"

No Comments

No Known Activations