INDEX

Explanations

references to loss functions in machine learning contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

tagHelperRunner

-0.56

fjspx

-0.53

 Fourier

-0.53

 fitness

-0.49

HtmlAttribute

-0.47

likleri

-0.46

 lược

-0.45

nocache

-0.43

parral

-0.43

Fourier

-0.43

POSITIVE LOGITS

 labels

0.96

 Labels

0.89

 Classes

0.89

 label

0.88

 Label

0.85

labels

0.82

Label

0.82

ClassId

0.82

Class

0.81

 class

0.81

Activations Density 0.314%

references to loss functions in machine learning contexts

No Comments

No Known Activations