INDEX

Explanations

references to loss functions and their properties in machine learning

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

loss function

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 kasarigan

-0.88

rungsseite

-0.69

ніципа

-0.67

RTEX

-0.65

<unused28>

-0.64

<unused8>

-0.64

<unused43>

-0.64

bootstrapcdn

-0.64

<unused74>

-0.64

<unused3>

-0.63

POSITIVE LOGITS

loss

0.43

 loss

0.36

Loss

0.35

losses

0.33

hilangan

0.31

 Loss

0.31

Contracts

0.31

 cost

0.30

 desarrollado

0.30

 cebo

0.29

Activations Density 0.016%

references to loss functions and their properties in machine learning

loss function

No Comments

No Known Activations