INDEX

Explanations

technical terms and jargon related to programming, particularly in system and software contexts

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 degrés

-0.33

../../../

-0.32

@[+][

-0.32

URBANA

-0.32

 remplacement

-0.32

浓

-0.31

Чита

-0.31

ChangeEvent

-0.31

 lamang

-0.31

şe

-0.30

POSITIVE LOGITS

featureID

0.65

ագրություններ

0.61

TypedDataSet

0.57

MigrationBuilder

0.54

RegressionTest

0.54

ViewFeatures

0.52

 initComponents

0.52

queryInterface

0.52

 Wikimedijinoj

0.52

StructEnd

0.50

Activations Density 0.200%

technical terms and jargon related to programming, particularly in system and software contexts

No Comments

No Known Activations