INDEX

Explanations

references to depth or profound concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 19-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_19/width_16k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.19.hook_mlp_out

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

DoubleQuotes

-0.62

ButterKnife

-0.53

httphttps

-0.52

bootstrapcdn

-0.50

 Infórmanos

-0.48

 Rejo

-0.47

 suite

-0.47

Multiline

-0.46

occuper

-0.45

frain

-0.45

POSITIVE LOGITS

 <<<<<<<<<<<<<<

0.55

يكب

0.55

oredCriteria

0.52

smarty

0.52

GraphicsUnit

0.50

xFFFFFF

0.50

-/**

0.49

",{

0.49

Gente

0.48

OuterClass

0.47

Activations Density 0.125%

references to depth or profound concepts

No Comments

No Known Activations