INDEX

Explanations

references to the term "class" across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

Források

-0.64

 Ciencia

-0.54

imread

-0.52

 abito

-0.50

jahres

-0.49

ướ

-0.48

 rejoindre

-0.48

cinogenicity

-0.48

detta

-0.47

 hâte

-0.47

POSITIVE LOGITS

 Class

1.81

 class

1.72

class

1.41

 CLASS

1.39

Class

1.32

 classe

1.22

 clase

1.14

 Classes

1.13

CLASS

1.10

 classes

1.07

Activations Density 0.136%

references to the term "class" across various contexts

No Comments

No Known Activations