INDEX

Explanations

technical terms related to neural networks and deep learning

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

Tokens after cited works

np_acts-logits-general · gemini-2.0-flash

trained models

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

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

avadoc

-0.39

 Machine

-0.38

Czar

-0.38

adoc

-0.37

アイテム

-0.37

 tent

-0.37

 Golem

-0.36

dataSource

-0.35

 analog

-0.35

 Pieces

-0.35

POSITIVE LOGITS

 estekak

0.91

Tembelea

0.90

AutoScaleMode

0.88

 propOrder

0.82

HasForeignKey

0.81

 дописавши

0.81

########.

0.81

 ProtoMessage

0.80

 مشين

0.79

AddTagHelper

0.79

Activations Density 0.176%

technical terms related to neural networks and deep learning

Tokens after cited works

trained models

No Comments

No Known Activations