INDEX

Explanations

references to programming or development functions and methods

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

 tartalomajánló

-0.83

SourceChecksum

-0.75

 виправивши

-0.70

ContentAlignment

-0.65

Rohy

-0.63

parsedMessage

-0.62

TagMode

-0.62

Tikang

-0.59

 nonUne

-0.59

queryInterface

-0.59

POSITIVE LOGITS

IBOutlet

0.56

 ebenfalls

0.33

SequentialGroup

0.32

 levantar

0.31

Quiero

0.29

 IBOutlet

0.29

ạp

0.29

GOTREF

0.28

0.27

 znajdu

0.26

Activations Density 0.001%

references to programming or development functions and methods

No Comments

No Known Activations